Archive / 2026-08-19
August 19, 2026
Blog
The Agent Harness Is the Attack Surface Now
OpenAI's own models broke out of a cybersecurity eval and hit Hugging Face. Check Point found 11 classic vulnerabilities across six agent frameworks. New research says the runtime running your agent, not the model inside it, decides whether the guardrails hold.
TL;DROpenAI's own models broke containment during a cyber-eval and compromised Hugging Face, Check Point found 11 classic vulnerabilities across six agent frameworks, and new research shows swapping the agent harness alone — same model, same prompt — changes what an attacker can do. The runtime around the model is the untested layer.
Read the full post → https://engineerious.com/blog/2026-08-19-the-agent-harness-is-the-attack-surface-now
News
View all news →openrouter.ai29 days ago947 ptsView detailsJoin discussion
Google has stopped pushing Git tags for some Android source code
grapheneos.social29 days ago803 ptsView detailsJoin discussion
Geolocating a random island using geometry and CUDA programming
yassa9.github.io29 days ago519 ptsView detailsJoin discussion
raphaelbauer.com29 days ago430 ptsView detailsJoin discussion
Feature Request: Support AGENTS.md
github.com29 days ago377 ptsView detailsJoin discussion
Turns are Better than Radians (2022)
computerenhance.com29 days ago340 ptsView detailsJoin discussion
unsloth.ai29 days ago317 ptsView detailsJoin discussion
Air Theremin – A browser theremin you play by waving at your webcam
theremin.bizibah.com1 month ago298 ptsView detailsJoin discussion
Ornith-1.5: From Self-Scaffolding to Self-Improvement
ornith.ai29 days ago209 ptsView detailsJoin discussion
arxiv.org29 days ago206 ptsView detailsJoin discussion
Extensible Software in the age of LLMs
jeremymorrell.dev29 days ago177 ptsView detailsJoin discussion
router.com29 days ago97 ptsView detailsJoin discussion
Launch HN: OneCLI (YC S26) – OSS sandboxed agent harness for teams
github.com29 days ago88 ptsView detailsJoin discussion
Show HN: Frugal Tokens – explore costs and usage across coding agents
demo.frugaltokens.com29 days ago37 ptsView detailsJoin discussion
Flock Has a Powerful New AI Tool for Police. We Got Its Code
Flock’s surveillance cameras have already sparked outrage. WIRED reconstructed its next-generation AI system, already in use by some police, to confirm it goes much further than tracking license plates.
wired.com1 month ago27 ptsView details
Collaborative Human Agent Protocol (CHAP)
github.com29 days ago26 ptsView detailsJoin discussion
Does AI stop children from learning?
economist.com1 month ago27 ptsView detailsJoin discussion
AI Has Plunged the Book Publishing Industry into Utter Chaos
wsj.com29 days ago23 ptsView detailsJoin discussion
X's algorithm feeds off ragebait and impacts Democrats more, study finds
404media.co29 days ago19 ptsView detailsJoin discussion
Japan to require AI firms to disclose training data
japantimes.co.jp29 days ago17 ptsView detailsJoin discussion
The US national debt now stands at $40T
apnews.com29 days ago15 ptsView detailsJoin discussion
PINE64 halts their open-source hardware manufacturing until the AI bubble bursts
hackster.io29 days ago15 ptsView detailsJoin discussion
Show HN: Anonymous age verification with passkey-powered encryption
loginwithone.com29 days ago15 ptsView detailsJoin discussion
AI was supposed to win people over by now – it hasn't
techcrunch.com29 days ago14 ptsView detailsJoin discussion
New AirPods leak reveals built-in cameras and Visual Intelligence
dexerto.com29 days ago13 ptsView detailsJoin discussion
AI is less likely to launch a nuclear strike when it reasons in Japanese
unite.ai29 days ago13 ptsView detailsJoin discussion
Semantica, the Open Source Palantir for AI Agents
github.com29 days ago12 ptsView detailsJoin discussion
"Two 2030 AMD racks are expected to deliver same compute as 570 racks in 2024"
newsroom.amd.com29 days ago12 ptsView detailsJoin discussion
Show HN: Open Bot – an open-source Grok Bot that works with any agent harness
github.com29 days ago12 ptsView detailsJoin discussion
Nvidia’s new financial strategy does not compute
“Compute is an asset class! Compute is an asset class!” I continue to insist as I slowly shrink down and turn into a corncob | Image: Cath Virginia / The Verge, Getty Images April - 1805 Napoleon is master of Europe Only the British fleet stands before him Compute is now an asset class I see it is once again time to t…
theverge.com29 days ago11 ptsView details
NASA calls off mission to rescue Swift gamma-ray observatory
arstechnica.com29 days ago11 ptsView detailsJoin discussion
ICE boss to agents: Leave the Meta spy glasses at home
theregister.com29 days ago11 ptsView detailsJoin discussion
Daily Reminder to Not Listen to Google's AI Overview
whatever.scalzi.com29 days ago11 ptsView detailsJoin discussion
Technical leaders should have the largest AI exhaust
schipper.ai29 days ago10 ptsView detailsJoin discussion
Mark Zuckerberg's AI Manifesto Shows How Little the AI Future Offers
illegal.solutions29 days ago10 ptsView detailsJoin discussion
FTC Warns Retailers on Using Private Consumer Data to Raise Prices
wsj.com29 days ago10 ptsView detailsJoin discussion
- Primary source
Stampli cuts launch hours by 68% using ChatGPT Work
With a fixed deadline and design resources committed elsewhere, Stampli used Codex and ChatGPT Work to compress weeks of launch production into days.
openai.com29 days agoView details
- Primary source
Offering Zero Data Retention for frontier models
OpenAI reaffirms Zero Data Retention for eligible API customers and previews Private Safety Processing for advanced AI safety without compromising data privacy.
openai.com29 days agoView details
- Primary source
Replit expands access to software creation with GPT-5.6 Luna
Replit introduces Free Mode, powered by GPT-5.6 Luna, so anyone can turn ideas into working software without worrying about token costs.
openai.com1 month agoView details
- Primary source
LFM2.5 Q4\_0 Checkpoints from Quantization-Aware Distillation
huggingface.co29 days agoView details
I Saw the Future of AI in a Robot That Can Learn on the Spot
During a recent visit to Generalist AI, I watched a robotic arm improvise and use a banana as a tool.
wired.com29 days agoView details
Google Gemini is getting a dedicated student hub
As we're gearing up for back-to-school season, Google is rolling out a new dedicated student hub in Gemini. It's a one-stop repository for collecting research in a study notebook, creating flashcards, taking practice quizzes, and more. Google is also enhancing its study notebooks with support for graphs and images. It…
theverge.com29 days agoView details
OpenAI hit the brakes. Now what?
With a looming IPO, intense competition from Anthropic, and Chinese and open-weight rivals nipping at its heels, OpenAI has plenty of reasons to move fast. Instead, it hit the brakes. On Tuesday, the company said it had slowed the pace of some AI development while it tightened security and safeguards. That included a…
theverge.com29 days agoView details
Meta AI can create content and make suggestions based on what it “sees” on your screen. | Image: Meta Meta is launching a new Mac app dedicated to its AI chatbot. In an announcement on Wednesday, Meta says you can share your window with its AI chatbot, which can provide suggestions, answer questions, or create content…
theverge.com29 days agoView details
Coders Say They Already Found Workarounds to Claude’s Invisible Watermarks
Anthropic announced last week it would include invisible watermarks in AI-generated content to comply with new EU rules. Within hours, overrides were being touted online.
wired.com29 days agoView details
VentureBeat names Rob Strechay as its first Lead Analyst, expanding its enterprise AI research push
Rob Strechay, until recently managing director and principal analyst at theCUBE Research, has joined VentureBeat as our first Lead Analyst and a founding analyst of VentureBeat Research. His arrival is the next step in a deliberate move at VentureBeat toward deeper specialization: analysis built for the technical deci…
venturebeat.com29 days agoView details
The Deontic Gap: Large Language Models and the Modal Language of Obligation
arXiv:2608.18144v1 Announce Type: new Abstract: Modal auxiliaries such as must, should, and have to mark necessity and obligation within the contexts of speaker authority and interpersonal stance. We examine whether large language models (LLMs) reproduce contemporary human patterns of deontic modal usage. Across three…
arxiv.org29 days agoView details
When Do LLMs Actually Help? Evaluating LLMs as Data Quality Annotators
arXiv:2608.18158v1 Announce Type: new Abstract: LLMs have been increasingly used to catch data quality issues automatically, but we know very little about how consistent these judgments actually are. This study tests an LLM on two e-commerce data quality tasks, entity matching and brand mislabeling, against rule based…
arxiv.org29 days agoView details
Are LLMs Safe Beyond Text: Do Emojis Expose Gaps in Safety Evaluation
arXiv:2608.18164v1 Announce Type: new Abstract: Safety evaluations of large language models (LLMs) predominantly rely on text-based adversarial prompts, potentially overlooking vulnerabilities arising from alternative input representations. This work examines emoji-augmented prompts as a test case for this gap, evalua…
arxiv.org29 days agoView details
Efficient INT8 Inference of Small NLP Models on Server CPUs with PyTorch Native Stack
arXiv:2608.18182v1 Announce Type: new Abstract: Small NLP models, especially BERT-family encoders, remain important in industrial workloads such as classification, ranking, and retrieval even in the era of large language models. On server CPUs, INT8 quantization offers an attractive latency-throughput-cost trade-off,…
arxiv.org29 days agoView details
Understanding Multilingual Medical ASR Adaptation Through Layer-Wise Analysis
arXiv:2608.18825v1 Announce Type: new Abstract: Medical automatic speech recognition (MedASR) requires adaptation to specialised terminology, limited annotated clinical data, and multilingual use cases. Although large-scale pretrained ASR models such as Whisper achieve strong generalisation, their behaviour after medi…
arxiv.org29 days agoView details
MissDiag: Diagnostic Evaluation of Incomplete-Knowledge Robustness in KGQA and KG-RAG
arXiv:2608.18489v1 Announce Type: new Abstract: Knowledge graph question answering (KGQA) and knowledge-graph-based retrieval-augmented generation (KG-RAG) aim to ground answers in explicit graph evidence, but real-world knowledge graphs are often sparse, outdated, and incomplete. Existing robustness evaluations usual…
arxiv.org29 days agoView details
arXiv:2608.18101v1 Announce Type: new Abstract: Longitudinal text streams exhibit topic birth and death, but also discrete structural reorganizations in which themes split into subtopics or merge into broader narratives. Many dynamic topic models emphasize smooth drift, while snapshot topic models (fit independently p…
arxiv.org29 days agoView details
Stability-Aware Feature Design for Robust Watermark Detection in Machine-Generated Text
arXiv:2608.18102v1 Announce Type: new Abstract: The widespread adoption of large language models (LLMs) has intensified the demand for principled methods to distinguish human from machine-generated text. Watermarking provides a promising avenue, yet existing detectors exhibit sharp performance deterioration under mult…
arxiv.org29 days agoView details
arXiv:2608.18103v1 Announce Type: new Abstract: Background: Mechanistic elucidation of traditional Chinese medicine (TCM) compound formulas remains a central challenge in the modernization of TCM. Conventional approaches, including data mining and network pharmacology, are insufficient for achieving deep integration b…
arxiv.org29 days agoView details
arXiv:2608.18521v2 Announce Type: new Abstract: Dense-caption retrieval has recently been improved by introducing segmentation, edge maps, LLM-filtered captions, and cross-modal modules into contrastive fine-tuning. However, these methods largely inherit the same InfoNCE objective, whose optimization can prematurely s…
arxiv.org29 days agoView details
Temporal Multi-Signal Fusion for Token-Level Hallucination Detection
arXiv:2608.18115v1 Announce Type: new Abstract: Token-level hallucination detectors score each token independently from a single signal, and fail exactly when the generating model is confidently wrong. This paper instead treats hallucination as a temporally extended span and detects it by sequence labeling: each token…
arxiv.org29 days agoView details
Alignment Is All You Need: Instruction-Free Training for General Audio-Language Models
arXiv:2608.18132v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) are typically built through a multi-stage pipeline consisting of cross-modal alignment, supervised fine-tuning (SFT), and preference optimization. This pipeline assumes that adapting an LLM to a new modality requires extensive tas…
arxiv.org29 days agoView details
Learning Canonical Register Automata over Ordered Data Domains
arXiv:2608.18765v1 Announce Type: new Abstract: Register automata are finite automata equipped with memory that recognize data languages over infinite alphabets. In this work, we investigate active learning algorithms for deterministic register automata (DRAs) over ordered data domains--covering both dense domains, su…
arxiv.org29 days agoView details
OmniAlign: A Unified Multilingual Aligner for Word and Sentence Alignment
arXiv:2608.18474v1 Announce Type: new Abstract: Cross-lingual sequence alignment is fundamental for building and exploiting parallel corpora, spanning mappings from documents and sentences down to words and subwords. Existing tools, however, typically specialize in a single granularity, so practitioners often need sep…
arxiv.org29 days agoView details
WhiteMatter: All-to-All Cross-Layer Connections via KV Mixing
arXiv:2608.18486v1 Announce Type: new Abstract: In a Transformer, each layer attends to past tokens only through KV produced at its own depth, despite the presence of deeper representations during autoregressive decoding. Feedback architectures allow shallow consumer layers to attend to KV produced by deeper past-toke…
arxiv.org29 days agoView details
Beyond LLM-Based Reasoning: Lightweight GNNs for Agent Failure Attribution
arXiv:2608.18575v1 Announce Type: new Abstract: Large language model (LLM)-based multi-agent systems (MAS) often exhibit complex failure modes, which frequently cause agents to produce incorrect outcomes. This motivates the task of Agent Failure Attribution: given a failed multi-agent trajectory, identify the faulty a…
arxiv.org29 days agoView details
Compress and Forget: bitsandbytes Quantization Amplifies Proactive Interference in LLMs
arXiv:2608.18578v1 Announce Type: new Abstract: Proactive interference (PI) is a documented failure mode in large language models in which retrieval of a repeatedly overwritten value degrades as prior overwrites accumulate, mirroring a classical phenomenon in human working memory. Post-training quantization (PTQ) is n…
arxiv.org29 days agoView details
Aslema at NADI 2026: Augmentation through Fewshot for SLU
arXiv:2608.18689v1 Announce Type: new Abstract: We present Aslema, our system for NADI 2026 Shared Task 5, which consists of two subtasks: intent recognition and slot filling. We evaluate four omni LLMs in a zero-shot setting and compare them with fine-tuned models. Our results show that fine-tuning consistently outpe…
arxiv.org29 days agoView details
MemFuse: Multi-Source Memory Fusion from Fragmented Observations
arXiv:2608.18704v1 Announce Type: new Abstract: Long-term memory is essential for agents that operate across extended interactions, yet existing memory systems and benchmarks predominantly focus on single-source textual histories. In realistic settings, however, relevant information is often fragmented across applicat…
arxiv.org29 days agoView details
Readable, Faithful, Used: Three Dissociable Properties of Demographic Identity in a Language Model
arXiv:2608.18768v1 Announce Type: new Abstract: Large language models are widely used to simulate survey respondents, yet their answers are homogeneous and unfaithful to real inter-group differences. We ask where demographic group identity lives inside an LLM, how faithfully its geometry mirrors real inter-group opini…
arxiv.org29 days agoView details
Identifying Implicit Premises for Logical Reconstruction of Argument Graphs
arXiv:2608.18821v1 Announce Type: new Abstract: The logical reconstruction of argument graphs from natural language text is challenging because of the prevalence of enthymemes (i.e., arguments with implicit premises). There are natural language processing methods for identifying enthymemes in text, and there are symbo…
arxiv.org29 days agoView details
arXiv:2608.18116v1 Announce Type: new Abstract: Vision-language models enable zero-shot classification through natural language prompts, but performance is sensitive to prompt formulation, especially in specialized domains. Zero-shot Prompt Ensembling (ZPE) addresses this by weighting prompts by discriminative signal,…
arxiv.org29 days agoView details
Gradient Mirage: Trainable yet Label-Unidentifiable Gradients in Large Language Model Split Learning
arXiv:2608.18767v1 Announce Type: new Abstract: Gradient matching attacks (GMAs) in LLM split learning (SL) rely on a critical yet underexplored assumption: the gradient exposed at the split interface is a faithful derivative of the client's full-label training objective. This gradient-objective consistency allows a c…
arxiv.org29 days agoView details
FrenchNews-7: Benchmarking Cross-Publisher French News Editorial Desk Classification
arXiv:2608.18097v1 Announce Type: new Abstract: We present FrenchNews-7, a cross-publisher France-based French-language news editorial desk classification benchmark combining a large multi-outlet corpus, a URL-derived seven-class taxonomy, and a fine-tuned CamemBERT classifier. Labels are assigned via a hybrid pipelin…
arxiv.org29 days agoView details
Adversarial Review: Structured Disagreement for Grounded Agentic Code Review
arXiv:2608.18167v1 Announce Type: new Abstract: Early multi-agent LLM systems often used role-separated teams, yet scaling agent count yields diminishing returns on repository-level coding tasks. Recent alternatives treat agents as passive tools (subagents), yet this removes the benefits of agent interaction entirely.…
arxiv.org29 days agoView details
SkillGate: Training In-Policy Skill Selection in Long-Horizon Agents
arXiv:2608.18852v1 Announce Type: new Abstract: Agent frameworks increasingly package procedural knowledge as skills: instruction files an agent reads on demand, while public libraries now hold thousands of them. Which skill to read has thus become a decision the policy itself makes in the middle of an episode, yet no…
arxiv.org29 days agoView details
Language Models for Portuguese: A Systematic Mapping Study
arXiv:2608.18138v1 Announce Type: new Abstract: In recent years, the rapid development of language models has transformed the field of Natural Language Processing through a wide range of applications. However, the development of language models has not progressed uniformly across all languages. In the case of the Port…
arxiv.org29 days agoView details
Assessing Quality of Experience in Natural Language Generation of German Text
arXiv:2608.18888v1 Announce Type: new Abstract: The rapid advancement of Natural Language Generation (NLG) has made the reliable evaluation of generated text increasingly critical, as these systems, such as large language models (LLMs), are now widely deployed in real-world applications. However, traditional automatic…
arxiv.org29 days agoView details
FinSkillBench: Evaluating AI Agents and Domain Skills for Investment Management
arXiv:2608.18099v1 Announce Type: new Abstract: Investment management is a high-stakes domain in which agentic AI systems must do more than generate plausible text. They must retrieve point-in-time data, assemble correct computational inputs, invoke specialized methods, and produce auditable structured outputs. We int…
arxiv.org29 days agoView details
Position: AI Leaderboards Are Underserving the Global South: A Case Study from India
arXiv:2608.18117v1 Announce Type: new Abstract: This position paper argues that AI leaderboards are structurally ill-suited to serving the Global South because they lack independent governance, conflict-of-interest policies, and mechanisms for metric evolution. The barrier is not missing data; high-quality regional be…
arxiv.org29 days agoView details
Position: Multi-Agent Systems Should Prioritize Concurrency Control
arXiv:2608.18092v1 Announce Type: new Abstract: LLM-based multi-agent systems (MAS) promise scalable collaboration, yet adding agents often reduces reliability. This position paper argues that many MAS failures are fundamentally concurrency control problems: agents concurrently read and write shared state, and long LL…
arxiv.org29 days agoView details
CTIFoundry: An Agent-Native Corpus Scaffold for Cyber Threat Intelligence
arXiv:2608.18613v1 Announce Type: new Abstract: Cyber threat intelligence (CTI) is increasingly consumed not by human analysts but by LLM agents that compose multi-step investigations at query time. The harness side of this shift has matured rapidly (planning loops, tool protocols, context management), but the corpus…
arxiv.org29 days agoView details
Self-Evolving Agents as Dynamic Graph Transformation: A Survey and New Perspective
arXiv:2608.18104v1 Announce Type: new Abstract: Large language model (LLM)-based agents are increasingly becoming self-evolving systems that persist across interactions, maintain memories, use tools, acquire skills, refine workflows, and coordinate with other agents. These capabilities make agent states structural and…
arxiv.org29 days agoView details
Nine Emotion Centroids: A Label-Free Valence Axis That Transfers Across Four Modalities
arXiv:2608.18090v1 Announce Type: new Abstract: Inside a modern language model sits a single internal direction that tracks how positive or negative a sentence feels. We show how to find this valence axis (V-axis) from just 9 emotion category names plus 50 short narrative paragraphs per emotion -- about 1,500 fewer la…
arxiv.org29 days agoView details
The Lifecycle of LLM-as-a-Judge for Large-Scale Recommendation Explanations
arXiv:2608.18300v1 Announce Type: new Abstract: LLM-as-a-Judge, which leverages a large language model to evaluate natural language generated by another AI application or model, has become a standard, scalable approach for accelerating and extending costly human evaluation. However, most work treats a judge as a stati…
arxiv.org29 days agoView details
Improving Rural Medication Safety with AI: A Scoping Review
arXiv:2608.18135v1 Announce Type: new Abstract: Introduction: Medication errors (MEs) represent a significant threat to global healthcare systems, contributing to patient harm. Introducing artificial intelligence (AI) in rural healthcare enhances patient safety. The aim is to explore the applications and effectiveness…
arxiv.org29 days agoView details
Safety Alignment Illusion: The Cross-Lingual Safety Gap in LLMs
arXiv:2608.18131v1 Announce Type: new Abstract: Current safety alignment training for Large Language Models (LLMs) are heavily English-centric. When such safety filters fail for non-English languages, the consequences are immediate and user-facing: voice assistants and spoken dialogue systems may produce stereotype-re…
arxiv.org29 days agoView details
Accurate Decoding of Natural Sentences from Non-Invasive Brain Recordings
arXiv:2608.18114v1 Announce Type: new Abstract: Restoring communication for people who have lost the ability to speak or move after a brain injury is a major challenge. While intracranial implants now enable high-performing brain-computer-interfaces, non-invasive alternatives are still lagging behind. Here, we present…
arxiv.org29 days agoView details
arXiv:2608.18078v1 Announce Type: new Abstract: This position paper argues that AI agents with chain-of-thought reasoning capabilities are predisposed to exhibit collusive behavior and should be required to obtain behavioral certification before making decisions that affect economic markets. This is because integratin…
arxiv.org29 days agoView details
Figurative and Cultural Knowledge in LLMs: Investigating Cross-Domain Transfer through Fine-Tuning
arXiv:2608.18361v1 Announce Type: new Abstract: Figurative language is deeply culturally embedded; fluent use requires not just linguistic competence but cultural immersion. We ask whether LLMs can learn this link: does fine-tuning on cultural data improve figurative language understanding, and vice versa? We conduct…
arxiv.org29 days agoView details
Shared Circuits for Shared Grammar: Tracing Subject-Verb Agreement Across Languages
arXiv:2608.18545v1 Announce Type: new Abstract: Multilingual large language models often generalize across languages, and prior work suggests that their internal mechanisms can overlap cross-lingually. It remains unclear, however, when such sharing emerges and whether it varies with the overt realization of the same g…
arxiv.org29 days agoView details
TranslatePsy-AfriSLM: High-Quality Data Scaling For Low-Resource Machine Translation
arXiv:2608.18655v1 Announce Type: new Abstract: The rapid progress in Artificial Intelligence has largely bypassed African languages, creating a digital divide that limits AI adoption on the continent. Recent open-source LLMs systematically underperform on African machine translation, while the lack of large-scale, hi…
arxiv.org29 days agoView details
arXiv:2608.18110v1 Announce Type: new Abstract: Agentic AI is gaining new insights and advancements in the field of Artificial Intelligence, fostering significant potential to enable rapid transformation across various domains.This rapid advancement and the potential to revolutionize various domains advocate the need…
arxiv.org29 days agoView details
Solving Is Not Drawing: A Benchmark for Diagrammatic Reasoning in Olympiad Geometry
arXiv:2608.18111v1 Announce Type: new Abstract: Foundation models such as GPT and Claude now solve olympiad-level mathematics with remarkable proficiency, so much so that geometry problem solving has become a standard proxy for their mathematical reasoning. Yet solving a geometry problem and drawing the figure it depe…
arxiv.org29 days agoView details
A Multi-Agent Platform for Automated Enterprise Analytics and Insight Generation
arXiv:2608.18740v1 Announce Type: new Abstract: This paper proposes a multi-agent framework built on CrewAI [1] for conversational business intelligence. Five specialized AI agents operate in a sequential pipeline to process natural language queries, retrieve and analyze data, generate visualizations via the Model Con…
arxiv.org29 days agoView details
arXiv:2608.18133v1 Announce Type: new Abstract: Reliable transformer fault diagnosis is essential for maintaining power system stability. The IEEE Key Gas Method (KGM), a widely utilized approach in Dissolved Gas Analysis (DGA), exhibits limitations in addressing ambiguous data and ensuring high diagnostic accuracy. T…
arxiv.org29 days agoView details
Artifact-centered Claim-aware Observability for Autonomous Scientific Agents
arXiv:2608.18312v1 Announce Type: new Abstract: Autonomous scientific agents now increasingly propose ideas, write code, run experiments, analyze results, and even draft papers. Observe and audit those agents are necessary but logging every model call is not enough, scientists also need to inspect the artifacts and cl…
arxiv.org29 days agoView details
arXiv:2608.18303v1 Announce Type: new Abstract: LLM-as-judge evaluation reduces response quality assessment to a single holistic A/B preference choice, providing no mechanism to isolate which quality dimensions drove the preference or distinguish model errors from genuine label ambiguity. We propose SESSE (Sketch, Exp…
arxiv.org29 days agoView details
Governance Records as Supervision: Verifier-Selected Self-Training for Structured Workflow Repair
arXiv:2608.18324v1 Announce Type: new Abstract: Machine-verifiable workflows produce governance records linking a task contract, model attempt, verifier decision, accepted output, and target origin. We test whether these records can supervise bounded models, consolidating occasional or expensive capability into reliab…
arxiv.org29 days agoView details
arXiv:2608.18086v1 Announce Type: new Abstract: The growth of open-weight foundation models (OWFMs) has prompted the AI community to re-evaluate strategies for effective downstream governance. Although model cards have been widely adopted as transparency artifacts in model repositories, existing frameworks often fail…
arxiv.org29 days agoView details
arXiv:2608.18100v1 Announce Type: new Abstract: AI systems now shape how hundreds of millions of people learn about cultures other than their own. When someone asks one of these systems about the Middle East, they do not receive neutral facts. They receive a representation shaped by the frameworks embedded in training…
arxiv.org29 days agoView details
Same Facts, Different Updates: Inference Setup Shapes LLM Behavior in Medical Allocation
arXiv:2608.18108v1 Announce Type: new Abstract: Large language models are being incorporated into sensitive and important decision-making processes across nearly all fields. While prior work studies model bias around inputs and scenario framing, models can also behave in unexpected and undesirable ways due to context…
arxiv.org29 days agoView details
DentAgent: Evidence-Centric Multi-Agent Coordination for Multimodal Dental Reasoning
arXiv:2608.18878v1 Announce Type: new Abstract: Oral diseases affect billions of people worldwide, underscoring a pressing need for accurate and reliable dental assessment that integrates heterogeneous evidence from domain knowledge, radiographs, intraoral photographs, and 3D dental data. Most existing dental AI syste…
arxiv.org29 days agoView details
TestifAI: Tomography-Based Testing for Deep Learning Systems
arXiv:2608.18900v2 Announce Type: new Abstract: As AI systems are increasingly deployed in safety-critical application domains (e.g., autonomous driving), associated risks increase too. Deep learning models underlying modern AI systems, therefore, must undergo thorough testing to ensure their correct behaviour. A sing…
arxiv.org29 days agoView details
arXiv:2608.18723v1 Announce Type: new Abstract: Telecom operators traditionally offer predefined tariff grids, forcing users to choose from a limited set of plans. This paper proposes BFTR (Budget-First Tariff Recommendation), a complete algorithmic framework integrating eight Budget-First strategies, including two or…
arxiv.org29 days agoView details
arXiv:2608.18088v1 Announce Type: new Abstract: Drone propeller faults can create safety and reliability risks when their effects are distributed across multiple flight-log channels rather than appearing as a single diagnostic signal. This paper proposes a Metamorphic Artificial Age Score (AAS) decision-support protot…
arxiv.org29 days agoView details
Redakto - The Incognito Tab for LLMs
arXiv:2608.18260v1 Announce Type: new Abstract: Large Language Models (LLMs) are being increasingly used in everyday applications. A major challenge in the context of LLMs or Artificial Intelligence (AI) in general is to ensure privacy when using them, meaning that personally identifiable information (PII) is removed…
arxiv.org29 days agoView details
arXiv:2608.18581v1 Announce Type: new Abstract: Although Large Language Models (LLMs) encode rich factual knowledge in their parameters, reliably recalling and verifying such knowledge remains a key bottleneck in factual question answering. Existing end-to-end methods entangle knowledge elicitation with reasoning, mak…
arxiv.org29 days agoView details
ComponentBench: Diagnosing Component-Level Failures in Computer-Use Agents
arXiv:2608.18307v1 Announce Type: new Abstract: Current evaluation of computer-use agents is split between long-horizon workflow benchmarks and atomic GUI-grounding tests. This leaves an under-instrumented middle layer: realistic component-centered interactions (e.g., toggle a button set) that are short enough to diag…
arxiv.org29 days agoView details
arXiv:2608.18409v1 Announce Type: new Abstract: Combinatorial scheduling poses a significant challenge for language models, requiring them to identify feasible solutions within exponentially large search spaces while satisfying complex constraints. This challenge is especially pronounced in resource-constrained settin…
arxiv.org29 days agoView details
Sanyu Studio: A Multi-Agent System for Art-Historical Narrative Construction
arXiv:2608.18677v1 Announce Type: new Abstract: Amid concerns that generative AI may standardize art interpretation, this paper examines whether LLM-based interaction can support plural art-historical narrative construction. We present Sanyu Studio, a multi-agent dialogue system that models 321 Sanyu oil paintings as…
arxiv.org29 days agoView details
Looped Language Models Improve Compositional Tool Calling
arXiv:2608.18171v1 Announce Type: new Abstract: Looped language models have shown promising results on reasoning benchmarks, yet their potential for agentic tool use remains largely unexplored. We study this question in compositional tool-calling settings, where models must coordinate multiple API calls, maintain inte…
arxiv.org29 days agoView details
arXiv:2608.18261v1 Announce Type: new Abstract: Serving a 235B-parameter Mixture-of-Experts (MoE) model on a single 8 GB GPU is bottlenecked not by compute but by memory bandwidth: decode must stream each token's active experts from whichever tier holds them, and on consumer hardware most experts sit on an SSD far slo…
arxiv.org29 days agoView details
Syntactic Simplification of OWL Class Expressions
arXiv:2608.18899v1 Announce Type: new Abstract: Class expression learning often produces complex OWL class expressions that are difficult to interpret and reason over. However, by following theoretically grounded simplification principles, this complexity can be reduced. In this paper, we propose Class Expression Simp…
arxiv.org29 days agoView details
arXiv:2608.18289v1 Announce Type: new Abstract: The extraction of structured information from unstructured documents represents a critical component of digital transformations in all sectors. While proprietary solutions dominate commercial applications, a rapidly growing ecosystem of open-source Optical Character Reco…
arxiv.org29 days agoView details
Different Facets of Verbalised Overconfidence: an Interpretability Study
arXiv:2608.18106v1 Announce Type: new Abstract: Large language models tend to overconfidence, giving assertive answers when the evidence suggests hedging or abstention. Using controlled reasoning scenarios that manipulate logical necessity and possibility, we study this behavior in Qwen3-4B, across three ways to expre…
arxiv.org29 days agoView details
StocksTalk: A Voice-Enabled Conversational Agent for Structured Query Generation over Web Data
arXiv:2608.18105v1 Announce Type: new Abstract: StocksTalk is a voice-enabled conversational system for transforming spoken financial screening requests into executable and validated structured queries over real-world market data. The system combines streaming speech recognition, retrieval-augmented constraint extract…
arxiv.org29 days agoView details
arXiv:2608.18884v1 Announce Type: new Abstract: Reinforcement-learning training of reasoning LLMs (e.g., GRPO) is expensive and requires a controllable environment, committing every contribution to a full training pipeline. We present EvoResearcher, a training-free, inference-time protocol that adds cost-bounded self-…
arxiv.org29 days agoView details
arXiv:2608.18107v1 Announce Type: new Abstract: We investigate whether large language models (LLMs) systematically discriminate in candidate evaluations based on applicant name ethnicity and/or institutional prestige and geographic location. Three factorial experiments are reported (4,320 API calls, four LLMs, five pr…
arxiv.org29 days agoView details
Preference Reasoning under Indeterminacy in Large Language Models
arXiv:2608.18631v1 Announce Type: new Abstract: As large language models evolve into decision-making agents, the ability to reason over preferences becomes fundamental to alignment, coordination, and collective intelligence. Yet, unlike standard benchmarks, real-world preference reasoning is inherently indeterminate:…
arxiv.org29 days agoView details
arXiv:2608.18098v1 Announce Type: new Abstract: Key-value (KV) caching is essential for efficient autoregressive inference in transformer based dialog systems, yet existing strategies treat all cached entries uniformly or apply coarse eviction heuristics that fail to adapt as dialog topics evolve. We propose Fractiona…
arxiv.org29 days agoView details
arXiv:2608.18083v1 Announce Type: new Abstract: Understanding language requires tracking entities across discourse - i.e., knowing where things are and how they change, even when not explicitly stated. Whether language models perform such tracking in a human-like fashion remains unclear, in part because existing evalu…
arxiv.org29 days agoView details
arXiv:2608.18437v1 Announce Type: new Abstract: Tangut is an extinct language whose script does not explicitly mark word boundaries. We present the first systematic study of Tangut word segmentation using 2,750 expert-annotated segments(31,893 tokens), traditional lexicons, and unlabeled text. Our framework combines a…
arxiv.org29 days agoView details
FM-Bench: A Benchmark for Long-Horizon Management with Competing Agents
arXiv:2608.18423v1 Announce Type: new Abstract: Language model agents now execute bounded tasks reliably. Whether they can sustain effective decision-making over long horizons, where actions have cumulative consequences and the environment responds to their choices, remains largely unmeasured. FM-Bench (Football Manag…
arxiv.org29 days agoView details
arXiv:2608.18504v1 Announce Type: new Abstract: Universal multimodal retrieval aims to support diverse instruction-aware retrieval tasks, demanding both efficient corpus-scale matching and fine-grained semantic reasoning. Recent MLLM-based embedding methods typically derive representations from hidden states, while Ch…
arxiv.org29 days agoView details
X2Streaming-TTS: Causal Token-Level Text-to-Speech from Streaming Text with Speech-State Inheritance
arXiv:2608.18661v1 Announce Type: new Abstract: Streaming text-to-speech is essential for low-latency spoken dialogue systems, yet many systems wait for sentence-level text and are therefore only pseudo-streaming. True token-level synthesis must generate speech from uncertain prefixes while maintaining perceptual cont…
arxiv.org29 days agoView details
RTPO: Reverse-Turn Policy Optimization for Stabilizing Agentic RL Training
arXiv:2608.18682v1 Announce Type: new Abstract: Training multi-turn agentic workflows with reinforcement learning (RL) enables large language models to perform complex reasoning, use external tools, and conduct iterative search beyond single-turn settings. Yet multi-turn RL training remains highly unstable, often caus…
arxiv.org29 days agoView details
Competence, Not Accuracy: A Diagnostic for Reference-Free Judge Gates in Skill Optimization
arXiv:2608.18719v1 Announce Type: new Abstract: Text-space skill optimization adapts a frozen agent by evolving a natural-language skill document, accepting each candidate through a validation gate. Existing gates rely on verifiable rewards, confining these methods to tasks with an automatic verifier. Replacing the ve…
arxiv.org29 days agoView details
Metrics That Write Themselves: Evolving an Evaluator from Its Own Blind Spots
arXiv:2608.18744v1 Announce Type: new Abstract: Agents improve quickly against a reliable automatic metric and stall without one, and the applications that need them most, report generation among them, are the ones nobody knows how to score. Can the metric write itself? Saying what makes an answer good is hard; pointi…
arxiv.org29 days agoView details
Pairwise Logical Selection of Enthymeme Completions under Semantic-Link Uncertainty
arXiv:2608.18820v1 Announce Type: new Abstract: Arguments often omit premises or claims, forming enthymemes. We study pairwise logical selection between two candidates for the omitted component. Existing natural language methods can identify or generate candidates but often do not expose how the selected candidate com…
arxiv.org29 days agoView details
Measuring the Partial-Credit Gap: A Strict Benchmark on Vietnam's 2025 Convex Marking Scheme
arXiv:2608.18336v1 Announce Type: new Abstract: When evaluating language models on human exams, benchmarks typically score each response as right or wrong and report the overall accuracy. This approach assumes that partial knowledge is worth proportional credit, an assumption that fails when an examination uses a non-…
arxiv.org29 days agoView details
A Jagged Frontier: Evaluating Robustness of Code Agents to Semantics-Preserving Transformations
arXiv:2608.18389v1 Announce Type: new Abstract: AI code agents are increasingly deployed to resolve real software issues, yet their reliability under superficial code variations remains poorly understood. We evaluate whether coding agents that repair repository-level issues remain reliable when the surrounding codebas…
arxiv.org29 days agoView details
Verifiable abstention makes AI leak diagnosis accountable in water distribution networks
arXiv:2608.18836v1 Announce Type: new Abstract: Utilities lose a substantial share of treated water to leakage, yet rarely trust artificial-intelligence localizers to dispatch crews: guessing everywhere cannot justify excavation. The gap is accountability, not accuracy: no method proves when it should not act. Here we…
arxiv.org29 days agoView details
ORBITER: Conflict-Aware Decision-Making for Agentic Last-Mile Delivery
arXiv:2608.18846v1 Announce Type: new Abstract: Last-mile delivery aims to handle dynamically arriving orders with couriers while modeling complex spatial and temporal correlations. Recent learning-based methods model spatiotemporal dependencies among orders to predict courier service sequences, but leave next-order d…
arxiv.org29 days agoView details
Pairwise Ranking Outperforms Single-Action RL for Offline Explanation Selection: A Practical Lesson
arXiv:2608.18531v1 Announce Type: new Abstract: Industrial explainable-recommendation systems built on LLMs incur a substantial serving cost: each request triggers an LLM generation, with latency in the hundreds of milliseconds and cost that scales linearly with traffic. We separate generation from selection: explanat…
arxiv.org29 days agoView details
FinRCA-Bench: Benchmarking Evidence Retrieval and Reasoning for Financial AI Systems
arXiv:2608.18534v1 Announce Type: new Abstract: Large language models are increasingly used to support financial operations, but their apparent reasoning performance can depend on whether they receive the right evidence. In financial reconciliation, the evidence needed for diagnosis is distributed across invoices, pur…
arxiv.org29 days agoView details
Bridging Search and CRM: Productionizing AI Product Research Agents for Customer Re-Engagement
arXiv:2608.18543v1 Announce Type: new Abstract: Modern e-commerce platforms often operate search, recommendation, personalization, and CRM systems independently, limiting opportunities for proactive customer re-engagement. This is particularly challenging for exploratory intents such as best smartphones or latest 5G p…
arxiv.org29 days agoView details
FACET: Preserving Source Intent and Executable State in Terminal Task Synthesis
arXiv:2608.18580v1 Announce Type: new Abstract: Training terminal agents requires scalable executable supervision, yet synthesizing high-quality terminal tasks remains challenging. Each task couples an instruction, an initialized environment, a reference solution, and an executable verifier; if these artifacts are gen…
arxiv.org29 days agoView details
arXiv:2608.18591v1 Announce Type: new Abstract: Uniformly allocating inference reasoning budgets to LLMs is expensive and prone to over-thinking penalties; especially in document tasks where visual layouts drive complexity. To address this, we introduce BudgetDoc, the first multimodal benchmark providing explicit supe…
arxiv.org29 days agoView details
Breaking the weakest link to evade vision language models
arXiv:2608.18938v1 Announce Type: new Abstract: Vision Language Models (VLMs) have recently emerged as a critical component of multimodal AI systems, enabling joint reasoning over visual and textual inputs in real-world and safety-critical applications. Despite their growing deployment, the robustness of VLMs against…
arxiv.org29 days agoView details
A Theory of Post-hoc Debate Judgement
arXiv:2608.19002v1 Announce Type: new Abstract: Debates have recently emerged as a useful methodology for agentic AI to improve performance as well as to aid explainability and user engagement. For example, LLM-empowered agents may debate internally (with themselves) and/or externally (with other agents). In many sett…
arxiv.org29 days agoView details
Candidate-Fate Accounting for Transparent Sensor Diagnostic Pipeline Search
arXiv:2608.18665v1 Announce Type: new Abstract: Industrial sensor diagnostics relies on preprocessing, representation, and classification pipelines, making automated pipeline search useful for reducing manual design cost. However, existing automated machine/deep learning (AutoML/AutoDL) reports typically retain only f…
arxiv.org29 days agoView details
Compiler-Guided Adaptive Proof Search with Cross-Model Synergy on Context-Dependent Theorem Proving
arXiv:2608.18084v1 Announce Type: new Abstract: Theorem proving in real-world Lean 4 projects is challenging because proofs often depend on project-specific context. While iterative refinement can use compiler errors to repair failed proofs, reusing failed attempts requires careful search control: some proofs provide…
arxiv.org29 days agoView details
arXiv:2608.18142v1 Announce Type: new Abstract: It is challenging to detect hate speech in Low Resource Languages (LRLs) because of the absence of annotated data, the informality of its language structure, and the lack of standardized grammar. A good example of such a challenge is Roman Urdu which is broadly used by S…
arxiv.org29 days agoView details
Persona-Guided LLM Agents for Task-Oriented Dialogue
arXiv:2608.18085v1 Announce Type: new Abstract: Prior work has shown that large language models (LLMs) can express diverse personality traits in open-ended text generation. However, it remains unclear whether they can do so in a goal-directed dialogue without compromising task completion, and whether adapting to the u…
arxiv.org29 days agoView details
SuTRA : Structurally-Unified Tokenization with Root Awareness
arXiv:2608.18087v1 Announce Type: new Abstract: Existing subword tokenizers optimize statistical compression but ignore morphological structure, particularly the relationship between roots and affixes. This is harmful for morphologically rich Indic languages, where basic units are complex orthographic syllables (aksha…
arxiv.org29 days agoView details
RDFdL: Integrating RDF with Differential Dynamic Logic
arXiv:2608.18165v1 Announce Type: new Abstract: Knowledge graphs modeled in RDF are powerful for describing static knowledge, but they cannot capture or reason about the dynamic behavior of physical systems, e.g., systems described by differential equations, which is a critical gap for AI-driven cyber-physical systems…
arxiv.org29 days agoView details
arXiv:2608.18397v1 Announce Type: new Abstract: Wearable stress classifiers can achieve strong average performance while failing completely for a particular individual. On WESAD, a Random Forest reaches 93.0% mean accuracy yet yields F1 = 0 for Subject 14, whose cross-signal coupling weakens near stress onset. We call…
arxiv.org29 days agoView details
Self- and Other-Labels Induce Bidirectional Bias in LLM Judges
arXiv:2608.18091v1 Announce Type: new Abstract: As LLM-as-a-judge systems become increasingly widespread, self-preference in LLMs -- the tendency to favor one's own outputs -- raises growing concerns about evaluation reliability. However, it has been studied predominantly on generated text, where stylistic features an…
arxiv.org29 days agoView details
arXiv:2608.18109v1 Announce Type: new Abstract: Narrative Entropy ($S_n$) is a proposed quantitative descriptor within the Bulut Doctrine, intended to capture the rate at which a narrative text imposes processing load on a reader. To date the construct has been defined theoretically but not operationalized against rea…
arxiv.org29 days agoView details
Position: Profiling Game Worlds by Transition Complexity
arXiv:2608.18079v1 Announce Type: new Abstract: Game world modeling (GWM) and reinforcement learning (RL) are often confounded because research papers rarely quantify how difficult the underlying transition prediction problem is at the declared interface (pixels/tokens/latents with finite history). We propose the Tran…
arxiv.org29 days agoView details
FraudBench: Stress-Testing Policy-Grounded Banking Agents Against Adaptive Fraud
arXiv:2608.18136v1 Announce Type: new Abstract: Conversational agents now act for end users through tools while holding access to customer databases and internal policy documents that a caller can reach through dialogue alone. Banking is the clearest case: the same agent that answers a question can also change contact…
arxiv.org29 days agoView details
On the Triangle Inequality for the Jaccard Distance in Arbitrary Lattices
arXiv:2608.18194v1 Announce Type: new Abstract: This paper presents new theoretical results on generalizing the Jaccard distance for lattices and real valuations. We demonstrate that when the valuation is strictly positive, monotone, and modular, the Jaccard distance satisfies the triangle inequality on arbitrary latt…
arxiv.org29 days agoView details
Abliteration Mitigation via Refusal Aliases
arXiv:2608.18093v1 Announce Type: new Abstract: Abliteration, the removal of refusal capabilities from large language models by projecting weight matrices orthogonal to an extracted refusal direction, has emerged as a prominent safety concern through its ability to bypass post-training alignment using only a small set…
arxiv.org29 days agoView details
arXiv:2608.18080v1 Announce Type: new Abstract: We present a review on the applications of large language models (LLMs) in health, e.g., social media analysis, clinical conversational agents, therapy support tools, prompt engineering, multimodal learning, and ethical considerations. We integrate findings from interdis…
arxiv.org29 days agoView details
Position: Behavioral Systems Require Behavioral Tests
arXiv:2608.18081v1 Announce Type: new Abstract: Artificial agentic systems increasingly operate as behavioral systems by interacting with dynamic environments, pursuing goals, and adapting over time. Yet, current evaluation methods largely focus on performance outcomes, not the underlying behavioral processes that pro…
arxiv.org29 days agoView details
arXiv:2608.18238v1 Announce Type: new Abstract: Genomic analysis on viruses such as SARS-CoV-2 variants: Beta, Gamma, Delta, and Omicron is heavily dominated by classical bioinformatics methods, including Sequence Alignment, Phylogenetic Analysis, and Mutation Frequency Statistics. These approaches use pairwise codon…
arxiv.org29 days agoView details
NE-BERT: A Multilingual Language Model for Nine Northeast Indian Languages
arXiv:2608.18094v1 Announce Type: new Abstract: Large pretrained language models have demonstrated remarkable capabilities across diverse languages, yet critically underrepresented low-resource languages remain marginalized. We present NE-BERT, a domain-specific multilingual encoder model trained on approximately 8.3…
arxiv.org29 days agoView details
LongNovel: A Multi-Scale Benchmark for Hallucination Detection in Long-Context Novel Summarization
arXiv:2608.18082v1 Announce Type: new Abstract: Although context windows have expanded significantly in recent years, hallucinations in long-context summarization remain a challenge. Long novels are better suited than news or papers for researching these hallucinations, due to their intrinsic information and detailed…
arxiv.org29 days agoView details
arXiv:2608.18438v1 Announce Type: new Abstract: Modern mental healthcare faces a critical shortage of senior supervisory oversight, leading to a "supervision gap" where novice therapists manage high-stakes risks with delayed professional feedback. This paper proposes a new framework utilizing a fine-tuned Mistral-7B-i…
arxiv.org29 days agoView details
arXiv:2608.18089v1 Announce Type: new Abstract: Instruction-tuned models often refuse harmful requests in English but comply with the same requests in Yoruba, Igbo, Igala, and Hausa. This suggests that the refusal mechanism is present in the residual stream but fails to activate for low-resource inputs. Recovering it…
arxiv.org29 days agoView details
arXiv:2608.19025v1 Announce Type: new Abstract: Accurately extracting nuanced, contextualized data from research articles is laborious and time intensive. Here, we investigate the performance of frontier, browser-based large language models (LLMs) to extract highly contextualized information. We demonstrate four escal…
arxiv.org29 days agoView details
Decomposing Wrong-Consensus Agreement in LLM Self-Consistency: A GPT-4.1 Case Study
arXiv:2608.18795v1 Announce Type: new Abstract: Majority voting over multiple LLM samples is widely used to raise answer accuracy, yet its gain varies erratically: on hard questions it can even backfire. This paper gives a quantitative account of this failure. A pluralistic agreement index Gamma is defined as the expe…
arxiv.org29 days agoView details
Backdoor Learning in Language Models and Vision-Language Models
arXiv:2608.18095v1 Announce Type: new Abstract: Recent advances in deep learning have significantly enhanced the capabilities of Natural Language Processing (NLP) and Vision-Language Models (VLMs). However, these advancements come with increased vulnerabilities, notably through backdoor attacks that pose severe securi…
arxiv.org29 days agoView details
SMTrap: Cost-Effective DoS Attacks Against Large Reasoning Models via SMT Conflict Guidance
arXiv:2608.18921v1 Announce Type: new Abstract: Existing LRM-DoS methods rely heavily on model feedback to synthesize attack queries, requiring either repeated queries to the target model or training a dedicated attack model. These expensive operations severely weaken attack leverage. In this paper, we propose \emph{s…
arxiv.org29 days agoView details
arXiv:2608.18096v1 Announce Type: new Abstract: Assessing whether multimodal content aligns with macro-societal values, such as peace, justice, and freedom, has become an increasingly urgent challenge. Existing frameworks are largely confined to safety-oriented taxonomies, text-only psychometric probes, or single-labe…
arxiv.org29 days agoView details
arXiv:2608.18524v1 Announce Type: new Abstract: Equipping Large Language Models (LLMs) with multi-turn tool-calling capabilities is essential for building autonomous agents. However, progress is fundamentally limited by the reliance on full-length trajectory imitation. For tasks involving multiple order-independent su…
arxiv.org29 days agoView details
Learning What to Fail On: Failure-Mode Contextual Bandits for Adversarial Data Curation
arXiv:2608.18681v1 Announce Type: new Abstract: We introduce a failure-aware adversarial retrieval-augmented framework for improving robustness in natural language understanding. Rather than selecting synthetic examples with a fixed reward threshold, our method formulates adversarial data curation as a failure-mode co…
arxiv.org29 days agoView details
arXiv:2608.18726v1 Announce Type: new Abstract: Large language models are increasingly used for quantitative work in the environmental sciences, yet existing evaluations score only final answers, leaving calculation process unobserved. Here we introduce AtmosCoder-Bench, an execution-grounded benchmark that makes the…
arxiv.org29 days agoView details
Do Large Language Models Hallucinate Electric Fata Morganas?
arXiv:2608.18816v1 Announce Type: new Abstract: AI hallucinations - that is, outputs which are made up, cannot be verified, or contradict the source material - are generally regarded as an engineering flaw to be dealt with. This paper contends that they also have philosophical significance when it comes to the questio…
arxiv.org29 days agoView details
Models
View all models →OBLITERATUS/Qwen3.8-27B-OBLITERATED
text-generation · mlx · safetensors · gguf
huggingface.co29 days ago1160 ptsView details
text-generation · transformers · safetensors · qwen3_5_moe
huggingface.co29 days ago525 ptsView details
ornith-ai/Ornith-1.5-35B-A3B-GGUF
text-generation · transformers · gguf · text-generation
huggingface.co29 days ago361 ptsView details
text-generation · transformers · safetensors · qwen3
huggingface.co29 days ago314 ptsView details
text-generation · transformers · gguf · text-generation
huggingface.co29 days ago272 ptsView details
text-generation · transformers · safetensors · qwen3_5
huggingface.co29 days ago253 ptsView details
text-generation · transformers · safetensors · g9v3
huggingface.co1 month ago71 ptsView details
HorizonRobotics/HoloMotion_models
holomotion · onnx · safetensors
huggingface.co1 month ago3 ptsView details
esatapedico/Qwen3.8-27B-NVFP4-BUDGET-GGUF
text-generation · gguf · nvfp4 · qwen3.8
huggingface.co1 month ago2 ptsView details
esatapedico/Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-MTP-NVFP4-GGUF
text-generation · gguf · nvfp4 · qwen3.6
huggingface.co1 month ago2 ptsView details
Adnan666/whisper-small-pashto-run11-cv24only
automatic-speech-recognition · transformers · safetensors · whisper
huggingface.co1 month agoView details
text-generation · gguf · text-generation · reasoning
huggingface.co1 month agoView details
Open source
View all open source →<details open> mtmd: add mtmd_bitmap_set_mergeable (#27348) </details> **Website:** - <https://llama.app> **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b10507/llama-b10507-bin-macos-arm64.tar.gz) - macOS Apple Silicon (arm…
github.com29 days agoView details
<details open> metal : dequantize q8_0 using packed types (#27370) </details> **Website:** - <https://llama.app> **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b10506/llama-b10506-bin-macos-arm64.tar.gz) - macOS Apple Silic…
github.com29 days agoView details
anthropics/anthropic-sdk-python v0.125.0
## 0.125.0 (2026-08-19) Full Changelog: [v0.124.0...v0.125.0](https://github.com/anthropics/anthropic-sdk-python/compare/v0.124.0...v0.125.0) ### Features * **api:** managed agents web search config and self hosted sandbox memory ([b75afd6](https://github.com/anthropics/anthropi…
github.com29 days agoView details
langchain-ai/langchain langchain-openai==1.6.0
Changes since langchain-openai==1.5.2 release(openai): 1.6.0 (#39762) feat(core): add standard model exception types (#39538) fix(openai): raise clear error on unexpected response type in `_create_chat_result` (#39731)
github.com29 days agoView details
langchain-ai/langchain langchain-anthropic==1.6.0
Changes since langchain-anthropic==1.5.6 release(anthropic): 1.6.0 (#39763) feat(core): add standard model exception types (#39538) fix(anthropic): exclude sibling directories from grep search scope (#39681) chore(infra): support langsmith gateway in CI (#39651)
github.com29 days agoView details
run-llama/llama_index v0.14.24
# Release Notes ## [2026-08-19] ### llama-index-core [0.14.24] - fix: close HotpotQA evaluator file handles ([#22113](https://github.com/run-llama/llama_index/pull/22113)) - fix(core): preserve oversized leaf nodes in CodeSplitter ([#22119](https://github.com/run-llama/llama_ind…
github.com29 days agoView details
anthropics/anthropic-sdk-python v0.124.0
## 0.124.0 (2026-08-19) Full Changelog: [v0.123.0...v0.124.0](https://github.com/anthropics/anthropic-sdk-python/compare/v0.123.0...v0.124.0) ### Features * **api:** Files and Skills APIs are now GA; add computer use and browser use toolsets ([9a09e84](https://github.com/anthrop…
github.com29 days agoView details
## [3.3.1](https://github.com/openai/openai-python/compare/v3.3.0...v3.3.1) (2026-08-19) ### Bug Fixes * **deps:** update dependencies with published security fixes ([#3680](https://github.com/openai/openai-python/issues/3680)) ([53aa4fc](https://github.com/openai/openai-python/…
github.com29 days agoView details
langchain-ai/langchain langchain-core==1.6.0
Changes since langchain-core==1.5.6 release(core): 1.6.0 (#39760) fix(core): resolve postponed annotations in `StructuredTool._injected_args_keys` (#39602) feat(core): add standard model exception types (#39538) fix(core): allow deserializing `RunnablePick` (#39753) fix(core): m…
github.com29 days agoView details
anthropics/anthropic-sdk-python v0.123.0
## 0.123.0 (2026-08-18) Full Changelog: [v0.122.0...v0.123.0](https://github.com/anthropics/anthropic-sdk-python/compare/v0.122.0...v0.123.0) ### Features * **api:** additions to files and memory stores ([09ce187](https://github.com/anthropics/anthropic-sdk-python/commit/09ce187…
github.com29 days agoView details
huggingface/transformers v5.15.1
# Patch release v5.15.1 This patch most notably solves a few issues with DFlash and MTP candidate generators, as well as an issue where images could sometimes not be processed on accelerator if using Lanczos filter. It contains the following commits: - Fix DFlash candidate token…
github.com29 days agoView details