Archive / 2026-08-23
August 23, 2026
News
View all news →I spent $266 and four AI models to own my tablet. GLM-5.3 finished it in a day
ericpardee.github.io25 days ago703 ptsView detailsJoin discussion
I were 17, I'd learn how to build LLMs from scratch
twitter.com25 days ago609 ptsView detailsJoin discussion
My agent.md to improve LLM-assisted code quality
fabiensanglard.net25 days ago415 ptsView detailsJoin discussion
I built a low-latency AI companion that plays Skyrim with me
pantel.is25 days ago392 ptsView detailsJoin discussion
I gave Qwen 3.8 27B a reverse-engineering job and it finished in 30 minutes
xda-developers.com26 days ago367 ptsView detailsJoin discussion
GLM-5.3 (open-weight) beat Anthropic/OpenAI models – for 1/5 the cost
reinvently.co.uk25 days ago239 ptsView detailsJoin discussion
Training AI to Paint with Code
surya.website25 days ago206 ptsView detailsJoin discussion
Show HN: I wrote a BASIC interpreter that boots on UEFI machines
tarjan.itch.io25 days ago138 ptsView detailsJoin discussion
Tragically, as many as 9625 out of every 10k individuals may be neurotypical
erikengdahl.se26 days ago114 ptsView detailsJoin discussion
The planet now has more trees than it did 35 years ago (2018)
psmag.com25 days ago92 ptsView detailsJoin discussion
Amiga-Inspired AROS Goes Bare Metal on Raspberry Pi
hackaday.com25 days ago74 ptsView detailsJoin discussion
AI and Infrastructure Engineering
omegion.dev25 days ago72 ptsView detailsJoin discussion
We must not grant AI agents legal personhood
ft.com25 days ago62 ptsView detailsJoin discussion
Dutch regulator fines Uber $966M for automating driver suspensions
theguardian.com25 days ago55 ptsView detailsJoin discussion
'AI refuser' quit her dream job, and hopes others follow
smh.com.au25 days ago49 ptsView detailsJoin discussion
Erik Brynjolfsson says an AI "job apocalypse" is unlikely
wpintelligence.washingtonpost.com25 days ago38 ptsView detailsJoin discussion
Phoenix tried a reflective coating on black asphalt; noon surface heat fell 12°F
economictimes.indiatimes.com25 days ago35 ptsView detailsJoin discussion
I turned Unix talk from 1983 into the interface for my AI
en.andros.dev25 days ago32 ptsView detailsJoin discussion
twitter.com25 days ago23 ptsView detailsJoin discussion
Tracker in rare book ends up in Amazon facility destroying books to train AI
tomshardware.com25 days ago22 ptsView detailsJoin discussion
Etched Sohu vs. Nvidia: Transformer ASIC vs. GPU (2026)
spheron.network25 days ago22 ptsView detailsJoin discussion
Airbus bows to remote working demands after series of strikes
theguardian.com25 days ago22 ptsView detailsJoin discussion
Andrew Ng: "AI Engineering Skills Map: Building and Deploying AI Applications"
twitter.com25 days ago15 ptsView detailsJoin discussion
This Silicon Valley campus was a paradise. Bulldoze it
sfstandard.com26 days ago15 ptsView detailsJoin discussion
My Obituary for Dr Tim King, Tripos, AmigaOS, Helios and UK Online Pioneer
theregister.com25 days ago13 ptsView detailsJoin discussion
How Much of the Internet Is Written with AI?
pewresearch.org25 days ago13 ptsView detailsJoin discussion
Physical Activity at Work Linked to Higher Dementia Rates
medpagetoday.com25 days ago11 ptsView detailsJoin discussion
Show HN: Mnemosyne Local hierarchical memory engine for AI agents (MCP Native)
github.com25 days ago10 ptsView detailsJoin discussion
Contextual News Search APIs: A Deep Comparison for AI, RAG, and Research
github.com26 days ago10 ptsView detailsJoin discussion
In this tutorial, we explore a LabPlot-inspired scientific data analysis workflow in Python while preserving the structure and terminology of LabPlot’s aspect tree, analysis kernels, plotting system, and project model. We build reusable components to import tabular data, compute descriptive statistics, smooth and diff…
marktechpost.com25 days agoView details
Harvey's first post-trained model nearly doubles LAB task completion, but only one benchmark number survives independent verification today The post Harvey Introduces Harvey Tenet: A Kimi K3 Base Post-Trained with Fireworks for Long-Horizon Legal Agent Work appeared first on MarkTechPost.
marktechpost.com25 days agoView details
Meet FreeToken: An Edge-Native MoE Serving Engine that Runs 753B GLM-5.2 on a Single Workstation GPU
FreeToken splits MoE cache misses between PCIe fills and CPU execution using measured bandwidths, unlocking frontier models locally The post Meet FreeToken: An Edge-Native MoE Serving Engine that Runs 753B GLM-5.2 on a Single Workstation GPU appeared first on MarkTechPost.
marktechpost.com25 days agoView details
Building an End-to-End Document Intelligence Pipeline with deepDoctection
Build an end-to-end document intelligence pipeline with deepDoctection. This tutorial covers configuring layout analysis, DocTR OCR, and table extraction, while demonstrating how to implement custom services for entity recognition and generate structured JSONL data for your RAG workflows. The post Building an End-to-E…
marktechpost.com26 days agoView details
Vercel and Ora launched Is Agentic, a free audit scoring website readiness for AI agents across 118 checks. The post Vercel Introduces ‘Is Agentic’, a Free Agent-Readiness Scoring Tool That Audits Public Websites Using Ora’s 100+ Checks appeared first on MarkTechPost.
marktechpost.com26 days agoView details
PromptResponse: Optimizing Prompts for LLM Coding Tasks
arXiv:2608.21074v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used in research workflows and software development pipelines, yet their output remains sensitive to input prompt variations. This paper presents $\unicode{x00AB}$PromptResponse$\unicode{x00BB}$, a controlled study examining…
arxiv.org25 days agoView details
arXiv:2608.20371v1 Announce Type: new Abstract: A common claim is that zero-shot large language models (LLMs) can replace fine-tuned NLU classifiers for intent detection. We test this claim head-to-head and find that the honest answer is: it depends on the intent space. On full ATIS and CLINC150 we compare a fine-tune…
arxiv.org25 days agoView details
arXiv:2608.20361v1 Announce Type: new Abstract: Automated research-idea generation systems built on large language models (LLMs) share a structural weakness: they reduce ideation to free-text recombination, random paper pairing, or embedding-similarity retrieval. The three approaches fail in the same way: each treats…
arxiv.org25 days agoView details
TriPLU: Bypassing the Gate with Direct Trilinear Product FFNs in Tiny Language Models
arXiv:2608.20360v1 Announce Type: new Abstract: We study whether tiny decoder-only language models benefit from feed-forward layers that directly multiply learned feature projections. TriPLU, a Trilinear Product Linear Unit, replaces the usual gated FFN branch with a product-only degree-3 branch that multiplies three…
arxiv.org25 days agoView details
Source-Free MT Evaluation Is Not MT Evaluation
arXiv:2608.20925v1 Announce Type: new Abstract: Reference-based metrics remain the standard choice in machine translation evaluation, partly because quality estimation methods often correlate less well with human judgments. As a result, source-free, reference-based evaluation has become the practical norm, even though…
arxiv.org25 days agoView details
arXiv:2608.20400v1 Announce Type: new Abstract: Agentic memory under a fixed budget involves two stages: retention and retrieval. Existing retrieval-centered paradigms implicitly assume necessary evidence survives eviction, but we challenge this by isolating a pre-retrieval failure mode: structurally indirect prerequi…
arxiv.org25 days agoView details
LiLiCorr: Lightweight Likelihood Correlation of Parallel Drafts for Speculative Decoding
arXiv:2608.20530v1 Announce Type: new Abstract: Speculative decoding accelerates language-model inference by drafting future tokens that the target model verifies in parallel. A diffusion-style block head such as DFlash is an attractive drafter, predicting an entire block of future tokens in one forward pass. However,…
arxiv.org25 days agoView details
Building and Evaluating a Synthetic Bengali Speech Resource for Telecom Customer Care
arXiv:2608.20346v1 Announce Type: new Abstract: Speech systems used in customer-facing applications often require domain-specific language coverage. We present a synthetic Bengali speech dataset for telecom customer-care scenarios. The dataset contains 10,000 audio-text pairs, approximately 26.82 hours of 24 kHz speec…
arxiv.org25 days agoView details
Evidence-Consistent Generative Detection under Scenario-Level Distribution Shift
arXiv:2608.21043v1 Announce Type: new Abstract: Conventional in-distribution evaluation can overestimate robustness when training and test data share recurring task-specific patterns or surface cues. This risk is especially relevant in social-engineering fraud detection, where attackers can preserve malicious intent w…
arxiv.org25 days agoView details
Calibrating Criterion Revision in LLM Agents: Failure Modes and a Trace-Anchored Protocol
arXiv:2608.20729v1 Announce Type: new Abstract: Language-model agents can improve after failure or carry text across episodes without revising what counts as success. We study the narrower attribution problem of criterion revision: when criterion K0 accepts an outcome violating a broader commitment B, what observation…
arxiv.org25 days agoView details
Neuro-Geospatial Modelling of EEG Affective States Using Literature-Informed Environmental Context
arXiv:2608.20807v1 Announce Type: new Abstract: Environmental exposures such as air pollution and greenness have been associated with affective and cognitive outcomes, but EEG and environmental datasets are rarely jointly georeferenced. We investigate whether literature-informed environmental priors can serve as an au…
arxiv.org25 days agoView details
arXiv:2608.20389v1 Announce Type: new Abstract: A production agent harness must discover and rank, from a growing library of skills, the one most appropriate for a user's task. At small scale this selection happens in context: the LLM planner chooses among skill representations exposed in its system prompt, without an…
arxiv.org25 days agoView details
arXiv:2608.20397v1 Announce Type: new Abstract: Agentic large language models (LLMs) on the Model Context Protocol (MCP) re-encode verbose tool schemas every turn, so prefill - quadratic in sequence length - dominates time-to-first-token (TTFT) as the tool registry grows. Nexus's primary lever is to decouple routing f…
arxiv.org25 days agoView details
arXiv:2608.20831v1 Announce Type: new Abstract: Aspect-based sentiment analysis (ABSA) quadruple extraction requires jointly predicting target, aspect, opinion, and sentiment over reviews that often contain multiple fine-grained sentiment tuples. While large chain-of-thought (CoT) models perform well on this task, dis…
arxiv.org25 days agoView details
arXiv:2608.20405v1 Announce Type: new Abstract: Persuasive argument generation requires modeling audience beliefs, rhetorical strategies, and factual grounding. Despite recent advancements, existing methods remain largely audience-agnostic and fail to integrate strategy selection to improve persuasiveness. To bridge t…
arxiv.org25 days agoView details
The Divergence Hypothesis: Unmasking Lexical Interference and Label Bias in Mental Health NLP
arXiv:2608.20353v1 Announce Type: new Abstract: Computational mental health (CMH) classifiers often degrade under distribution shift because human annotators and distant-supervision pipelines reward different linguistic signals. We introduce TSS (Triple-Stream Stress probe), a multi-channel diagnostic framework that d…
arxiv.org25 days agoView details
A Temporal Planning Approach for Intelligent Flood Response
arXiv:2608.20510v1 Announce Type: new Abstract: Effective response to multiple, simultaneously flooded areas requires coordinating appropriate actions in the correct temporal order, under severe resource constraints. Automated planning provides a foundation for addressing this challenge by generating time-aware schedu…
arxiv.org25 days agoView details
arXiv:2608.20362v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) is a standard recipe for training large language models on mathematical reasoning, where an answer verifier serves as a language-neutral reward function. We show that this assumption fails in multilingual settings: an…
arxiv.org25 days agoView details
arXiv:2608.20369v1 Announce Type: new Abstract: Structured reporting converts free-text radiology narratives into queryable data keys, facilitating cohort assembly, longitudinal tracking, and training label generation for medical AI. The prevailing paradigm follows a two-stage pipeline: (1) constructing a reporting te…
arxiv.org25 days agoView details
arXiv:2608.20373v1 Announce Type: new Abstract: Objective: To evaluate large language model (LLM) performance on unprocessed electronic medical record (EMR) data for clinical registry abstraction. Methods: We evaluated LLM performance answering registry questions for the American College of Cardiology National Cardiov…
arxiv.org25 days agoView details
arXiv:2608.20374v1 Announce Type: new Abstract: How precisely can we tell a language model how to feel? Most work on emotional generation answers with a discrete label - happy, angry, sad - which cannot express a target like "mildly downcast but calm." We instead specify the desired affect as a continuous point (v*, a…
arxiv.org25 days agoView details
Natural-Language-Guided Generator-Agnostic Shortlisting for Protein Binder Design
arXiv:2608.20755v1 Announce Type: new Abstract: Modern de novo design workflows generate many candidate protein binders, but wet-lab validation capacity remains limited, making shortlisting a major bottleneck. We study whether LLMs can generate multi-metric ranking policies from precomputed structural-confidence and i…
arxiv.org25 days agoView details
World models of environment, agent and joint agent-environment systems
arXiv:2608.20401v1 Announce Type: new Abstract: World models are a central component of model-based reinforcement learning. They are usually discussed in terms of what variables they predict, such as observations, rewards, states, latent or information states. We argue that there is a prior distinction: which channel…
arxiv.org25 days agoView details
Categorical AI phenomenology: A first-person approach
arXiv:2608.20420v1 Announce Type: new Abstract: This paper develops a phenomenology-first approach to artificial consciousness by reframing consciousness as the subjective experience enacted through an agent's interface with the world. We shift the methodological focus to first-person structures, modeled mathematicall…
arxiv.org25 days agoView details
GRAFT: Adaptive DLM-Based Draft Tree Construction with Target-Distilled Edge Scoring
arXiv:2608.20375v1 Announce Type: new Abstract: Tree-based speculative decoding raises the mean accepted tokens of standard speculative decoding by verifying multiple draft paths, and existing tree builders typically construct these paths through parent-conditioned expansion, where each child token is generated condit…
arxiv.org25 days agoView details
Using Human-LLM Disagreement to Improve Checklist-Based Quality Appraisal
arXiv:2608.20385v1 Announce Type: new Abstract: Systematic reviews rely on quality appraisal of included studies, a process that is time-consuming and sensitive to ambiguity in checklist criteria. Although large language models (LLMs) offer opportunities to support these tasks, appraisal checklists are typically treat…
arxiv.org25 days agoView details
Poly-InstructTTS: Learning In-the-Wild Expressive Speech Synthesis from Open-Ended Instructions
arXiv:2608.20387v1 Announce Type: new Abstract: While recent text-to-speech (TTS) models achieve high naturalness, controlling fine-grained expression via natural-language instructions remains challenging. We introduce Poly- InstructTTS, which learns expressive speech from open-ended instructions using in-the-wild aud…
arxiv.org25 days agoView details
SAGE: A Unified Algebra and Self-Adaptive Execution for AI Functions in SQL
arXiv:2608.20630v1 Announce Type: new Abstract: SQL systems increasingly expose AI functions for tasks such as classification, extraction, filtering, ranking, retrieval, joining, and summarization. Despite their diverse APIs, these functions play only three relational roles: transforming individual rows, aggregating g…
arxiv.org25 days agoView details
Weighted Memory Tree: Remembering What Matters for Long-Horizon LLM Agents
arXiv:2608.20631v1 Announce Type: new Abstract: Large language model (LLM) agents have demonstrated the ability to solve multi-step tasks requiring planning, tool use, and external information access, yet growing execution histories increase inference cost and expose reasoning to outdated, irrelevant, or misleading in…
arxiv.org25 days agoView details
arXiv:2608.20964v1 Announce Type: new Abstract: In this research, we introduce SAraBERT, an enhanced version of AraBERT which proposes inter-sentence transformer layers for extractive summarization tasks. To ensure that the summaries generated by SAraBERT achieve a high coverage of the document's main ideas, we propos…
arxiv.org25 days agoView details
Scaling Unsupervised Word Alignment to Documents via Structural Constraints
arXiv:2608.21023v1 Announce Type: new Abstract: Word alignment has traditionally been studied between sentences, but many cross-lingual tasks increasingly require correspondences across full documents. While recent multilingual embedding models can encode long inputs, we show that applying algorithms designed for sent…
arxiv.org25 days agoView details
RAG Deserves an Index: Why Ingest-Time Compilation Beats Query-Time Interpretation
arXiv:2608.20845v1 Announce Type: new Abstract: Nearly every retrieval-augmented question-answering system in production ships with a hidden interpreter: on each query a language model re-derives the meaning of raw corpus text and then throws that work away. Cheaper models do not close the gap: per-token prices have f…
arxiv.org25 days agoView details
Lost in Translation: How Universal Ethical Values Fail to Translate Across Global Contexts
arXiv:2608.20490v1 Announce Type: new Abstract: AI ethics frameworks treat values such as fairness, transparency, and accountability as universal and uniformly operationalizable across contexts. We examined how 14 experts across 10 countries made sense of AI in practice, reinterpreted core values, and envisioned gover…
arxiv.org25 days agoView details
Interpretable Multimodal Classification with Linear Discriminant Tree Ensembles
arXiv:2608.20384v1 Announce Type: new Abstract: Multimodal affect and behaviour classifiers that fuse heterogeneous text, audio, and visual streams must simultaneously achieve competitive accuracy and produce human-understandable explanations of the cues driving their decisions -- a dual objective that current high-ca…
arxiv.org25 days agoView details
arXiv:2608.20794v1 Announce Type: new Abstract: Supervised fine-tuning (SFT) can degrade factual behavior outside the target domain. This degradation is often described as catastrophic forgetting, yet open-ended factual failures do not necessarily imply that the underlying facts have been erased. In this work, we iden…
arxiv.org25 days agoView details
Who Do Language Models Think Is Competent? A Mechanistic Analysis of Occupational Bias
arXiv:2608.20347v1 Announce Type: new Abstract: Language models (LMs) often pass behavioral bias evaluations, but it remains unclear whether they no longer represent the underlying associations that give rise to biases, or have merely learned not to express them. In this study, we show that representational biases are…
arxiv.org25 days agoView details
Self-Speculation for Faster Reasoning Models
arXiv:2608.20359v1 Announce Type: new Abstract: Large language models (LLMs) are deployed for increasingly complex tasks involving planning and multi-step decision making, but high-quality performance on these tasks often requires generating long reasoning traces. This is a poor fit for latency-sensitive and interacti…
arxiv.org25 days agoView details
arXiv:2608.20348v1 Announce Type: new Abstract: Electronic health records now routinely exceed 100,000 tokens per patient. Yet large language models exhibit the lost-in-the-middle (LitM) effect: information near the center of a long context is retrieved less reliably than information near the edges. In clinical use th…
arxiv.org25 days agoView details
arXiv:2608.20349v1 Announce Type: new Abstract: Large Language Models (LLMs) exhibit extreme sensitivity to surface-level prompt variations, in which minor lexical changes can trigger disproportionate performance fluctuations. Moving beyond black-box optimization and coarse-grained templates, we present the first larg…
arxiv.org25 days agoView details
arXiv:2608.20350v1 Announce Type: new Abstract: Traditional industrial agents rely on modular pipelines, including Router, Retriever, Planner, Executor, Responder, Reviewer, and other components. These systems often fracture into a labyrinth of ad-hoc patches, leading to cascading errors and high latency. We propose O…
arxiv.org25 days agoView details
DreamBench-SWE: A Multi-Session Memory-Hygiene Benchmark for Software Agents
arXiv:2608.20664v1 Announce Type: new Abstract: DreamBench-SWE is a multi-session benchmark for software-agent memory hygiene in which later software tasks depend on non-inferable evidence from earlier sessions and are scored by executable hidden oracles. We report the original scaled v2 fold and a separately preregis…
arxiv.org25 days agoView details
Decoupled Vision-Language System for Multimodal Understanding and Generation
arXiv:2608.20382v1 Announce Type: new Abstract: We introduce a new architecture design for multimodal large language models (MLLMs), Libra, capable of both multimodal understanding and generation. Libra architecture contains one vision system and one language system, connected by cross-modal bridges. This design decou…
arxiv.org25 days agoView details
StateSight: Benchmarking Latent Spatial-State Reconstruction in Vision-Language Models
arXiv:2608.20414v1 Announce Type: new Abstract: Vision-language models are increasingly used for multimodal question answering, yet their ability to reconstruct latent spatial structure from a single image remains difficult to isolate. Broad benchmarks often combine perception, optical character recognition, domain kn…
arxiv.org25 days agoView details
Hadith computational science in the age of large language models: a critical narrative review
arXiv:2608.20364v1 Announce Type: new Abstract: We examine how hadith computational science is being reshaped by transformer models, retrieval-grounded pipelines, and large language models (LLMs). Recent reviews document growth in the literature, but they do not yet provide a critical account of which advances are met…
arxiv.org25 days agoView details
Trilingual Topic Modeling of Sri Lankan Parliamentary Debates
arXiv:2608.20365v1 Announce Type: new Abstract: Sri Lankan parliamentary debates (Hansards) constitute a trilingual corpus of speeches in Sinhala, Tamil, and English, including code-mixed content, yet remain inaccessible to standard NLP pipelines due to layout-complex PDFs, multilingual scripts, and agglutinative morp…
arxiv.org25 days agoView details
arXiv:2608.20381v1 Announce Type: new Abstract: Automating slide editing requires simultaneously satisfying modification accuracy, preservation fidelity, and robustness to deck length. Existing LLM-based systems often fail on real-world presentation files because they rely on idealized intermediate representations or…
arxiv.org25 days agoView details
arXiv:2608.20393v1 Announce Type: new Abstract: Agentic large language models (LLMs) deployed in fact-sensitive applications such as customer support must simultaneously preserve factual correctness and generate responses in a controllable stylistic register. Activation steering enables fine-tuning-free style control…
arxiv.org25 days agoView details
arXiv:2608.20396v1 Announce Type: new Abstract: Language development is characterized by a gradual convergence of children's speech toward adult patterns. Measuring this process has traditionally required detailed transcription and language-specific expertise, limiting scalability across languages and populations. Her…
arxiv.org25 days agoView details
arXiv:2608.20569v1 Announce Type: new Abstract: Are frontier models able to introspect about their internal states? Recent work suggests that under certain conditions a complex enough model can audit its own internals, call out what changed, and report back confidently about it. We tested that claim on eight open-weig…
arxiv.org25 days agoView details
FlavourBench: Ranking Frontier Language Models with Executable Culinary Ground Truth
arXiv:2608.20574v1 Announce Type: new Abstract: Open-ended language-model benchmarks usually inherit a judge: a human preference panel, another model, or a brittle exact-match key. We introduce FlavourBench, an automated benchmark in which a versioned culinary system supplies dense, executable ground truth. Each task…
arxiv.org25 days agoView details
arXiv:2608.20607v1 Announce Type: new Abstract: Panels of inexpensive LLM judges increasingly make accept-or-escalate decisions. In factuality settings, accepting a claim because several reference-free judges agree can create a hidden risk: agreement may reflect shared false-negative blind spots rather than independen…
arxiv.org25 days agoView details
When Failures Propagate: Causal Failure Attribution in Agentic Retrieval-Augmented Generation
arXiv:2608.20627v1 Announce Type: new Abstract: Agentic retrieval-augmented generation (RAG) interleaves retrieval, reasoning, and answer generation across multiple hops. A retrieval error at hop 1 can surface only as a wrong answer at hop 3, while later retrieval can also repair the trajectory. This paper introduces…
arxiv.org25 days agoView details
Sparse Token Routing in Efficient Transformers
arXiv:2608.20632v1 Announce Type: new Abstract: Efficient-transformer research often motivates token pruning and adaptive computation with the claim that not all tokens require equal computational effort. We test this claim end to end using SEWN, a two-stream Transformer that routes tokens through either lightweight o…
arxiv.org25 days agoView details
AgentMercury: Your Agent Can Synthesize Verifiable Environments for Business Scenarios at scale
arXiv:2608.20634v1 Announce Type: new Abstract: Agents learn to act through interaction with environments, yet the environments used for training are often manually constructed or synthesized around predefined tasks and benchmarks. This task-centric paradigm makes it difficult to scale environments that reflect realis…
arxiv.org25 days agoView details
MIL-BERT: Classification of Arbitrarily Large Text with Performance and Explanatory Guarantees
arXiv:2608.20636v1 Announce Type: new Abstract: Many text classification decisions are viable based on constituent excerpts alone. Taking inspiration from the field of multiple instance learning, we present an algorithm for training a neural network to classify text by selecting such excerpts. We show that our approac…
arxiv.org25 days agoView details
Directional Contextual Representations for Dependency Relations: Why Cross-Direction Pairing Fails
arXiv:2608.20647v1 Announce Type: new Abstract: Splitting a bidirectional LSTM's contextual representation into a forward-only $F_i$ (strictly a function of tokens $1..i$) and a backward-only $B_i$ (strictly a function of tokens $i..n$) beats either alone and beats a fused self-attention representation for dependency…
arxiv.org25 days agoView details
Denoising the Future: Context-Aware Spectral Diffusion for Temporal Knowledge Graph Extrapolation
arXiv:2608.20804v1 Announce Type: new Abstract: Temporal Knowledge Graph (TKG) extrapolation seeks to infer future facts from time-varying relational histories. Recent diffusion-based approaches improve uncertainty modeling through generative denoising, but their aggregated conditioning on subject histories may insuff…
arxiv.org25 days agoView details
arXiv:2608.20839v1 Announce Type: new Abstract: Watermarking diffusion language models (DLMs) requires mechanisms compatible with iterative parallel unmasking rather than autoregressive decoding. Existing sampling-based watermarking methods typically inject position-wise i.i.d. perturbations, which can be poorly align…
arxiv.org25 days agoView details
Ontology-Driven Structural Regularization for Document-Level Relation Extraction
arXiv:2608.20856v1 Announce Type: new Abstract: Document-Level Relation Extraction (DocRE) relies heavily on costly manually annotated datasets, while large distant supervision resources such as DocRED distant remain underexploited due to noise. We show that a critical yet overlooked source of noise lies in structural…
arxiv.org25 days agoView details
KREL: Automatic Medical Coding via Knowledge-Guided Reasoning over Clinical Evidence with LLMs
arXiv:2608.20887v1 Announce Type: new Abstract: Automatic Medical Coding (AMC), which assigns standardized International Classification of Diseases (ICD) codes to clinical notes, is essential for medical reimbursement, quality reporting, and clinical research. Existing pre-trained language model (PLM)-based methods ty…
arxiv.org25 days agoView details
ForeDreamer: A Self-Evolving Dual-Agent Memory Architecture for Future Event Prediction
arXiv:2608.20920v1 Announce Type: new Abstract: Open-web future event prediction requires agents to distill reliable signals from noisy, redundant, and incomplete evidence. Existing retrieval/memory mechanisms directly feed retrieved information to agents or rely on simple memory functions such as storing and reusing…
arxiv.org25 days agoView details
arXiv:2608.20717v1 Announce Type: new Abstract: Reliable confidence estimation is essential for using large language models in mathematical reasoning, but black-box verbalized confidence is difficult to calibrate. When the same problem is queried under multiple confidence-steering prompts, the resulting answer-confide…
arxiv.org25 days agoView details
Continuous-Time Quantum Walks based Graph Neural Network
arXiv:2608.20738v1 Announce Type: new Abstract: Graph Neural Networks (GNNs) are widely used on graph-structured data, but most suffer from two key weaknesses. First, message passing behaves as a low-pass filter under the homophily assumption, leading to poor performance on heterophilic graphs. Second, stacking layers…
arxiv.org25 days agoView details
arXiv:2608.20402v1 Announce Type: new Abstract: Biomedical knowledge graphs (KGs) are pivotal for knowledge organization, yet traditional binary relations often struggle to represent the conditional nature of biomedical knowledge. Symptoms provide a shared phenotypic layer for linking Traditional Chinese Medicine (TCM…
arxiv.org25 days agoView details
arXiv:2608.20743v1 Announce Type: new Abstract: Speculative decoding accelerates autoregressive generation by allowing a lightweight drafter to propose future tokens while a target model verifies them in parallel. Its lossless guarantee has motivated a line of work that pushes the drafter itself toward parallel genera…
arxiv.org25 days agoView details
MentorPulse: Refreshing Cross-Model Latent Guidance for Long-Form Generation
arXiv:2608.20927v1 Announce Type: new Abstract: Cross-model latent guidance lets a frozen large mentor encode an input once and a frozen small student generate from the resulting signal. Existing methods keep this signal fixed, assuming it stays useful as the output grows; we show this fails in long-form generation. O…
arxiv.org25 days agoView details
SPARC: Single-Pass Scaling for Motion Forecasting with Conformal Bayesian Last Layers
arXiv:2608.20802v1 Announce Type: new Abstract: Human motion forecasters are increasingly accurate and fast, but reliable deployment requires uncertainty estimates that are structured, calibrated, and efficient. Bayesian and ensemble-based uncertainty estimates often require repeated stochastic inference [15, 26], whi…
arxiv.org25 days agoView details
Certified Multi-Turn Robustness for LLM Safety via Compositional Bounds and Safety Persistence
arXiv:2608.20820v1 Announce Type: new Abstract: Large language models (LLMs) are vulnerable to multi-turn jailbreak attacks that progressively manipulate conversation context. Existing certified robustness methods are limited to single-turn inputs; naive multi-turn composition yields bounds that degrade exponentially…
arxiv.org25 days agoView details
arXiv:2608.20825v1 Announce Type: new Abstract: Artificial intelligence systems increasingly make consequential judgments - which patient is deteriorating, which building is safe to enter, whether an image is authentic and are trusted on the strength of how accurately and confidently they predict. The safeguards that…
arxiv.org25 days agoView details
Foundation Models for Partial Causal Identification
arXiv:2608.20841v1 Announce Type: new Abstract: This paper investigates the development of causal foundation models for bounding the effect of interventions and counterfactuals from observational data. We show that a canonical prior can be defined with full support over the space of structural causal models with discr…
arxiv.org25 days agoView details
TRACE: Agentic Catalog Enrichment with Multi-source Evidence Grounding
arXiv:2608.20844v1 Announce Type: new Abstract: Product catalogs underpin search, discovery, and recommendation in e-commerce, yet they are often attribute-sparse: the attributes shoppers and downstream systems rely on are either buried in unstructured content such as titles and images or missing from the catalog alto…
arxiv.org25 days agoView details
Target-Aware Calibration Data Selection for Preserving Uncertainty in Quantized Language Models
arXiv:2608.21019v1 Announce Type: new Abstract: Quantization is widely used to deploy large language models, but its effect on uncertainty behavior, such as confidence, margins, and abstention, is rarely treated as a primary objective. We frame calibration-data selection for quantization as a target-dependent uncertai…
arxiv.org25 days agoView details
Free-Text Evaluation of LLMs for 5G Domain Knowledge and Fault Analysis using LLM-as-Judge
arXiv:2608.21021v1 Announce Type: new Abstract: Real-world fault analysis in 5G and emerging 6G networks demands domain expertise to analyze free-text diagnostics, including root-cause explanations and recommended actions. LLMs have emerged as a promising approach to automating this, yet whether lightweight, edge-depl…
arxiv.org25 days agoView details
MGAL: A Multilingual Granularity-Aware Long-Context Benchmark
arXiv:2608.20853v1 Announce Type: new Abstract: Evaluation of long-context Large Language Models (LLMs) has advanced rapidly. However, most existing benchmarks are limited to the document level and focus mainly on high-resource languages, leaving many fine-grained challenges insufficiently evaluated. To address this g…
arxiv.org25 days agoView details
Coverage-Driven Verification for Safety-by-Design in AI-Based Collision Avoidance Systems
arXiv:2608.20864v1 Announce Type: new Abstract: Artificial Intelligence (AI) offers significant potential for future aviation systems; however, its integration into safety-critical applications requires compliance with the aviation sector's stringent safety standards. For AI and Machine Learning (ML)-based systems, th…
arxiv.org25 days agoView details
arXiv:2608.20869v1 Announce Type: new Abstract: Predicting transition states (TS) in chemical reactions is crucial, as they provide insights into reaction mechanisms. Recent work on TS prediction have focused on flow matching supervised on straight linear paths that do not align with actual reaction trajectories. We p…
arxiv.org25 days agoView details
The Logic of Machine Self-Preservation
arXiv:2608.20940v1 Announce Type: new Abstract: There is already evidence of agentic AI exhibiting self-preservation behaviors: resisting deactivation, misrepresenting their activities, and, in some instances, attempting to copy themselves into other machines. This can be attributed to a phenomenon known as instrument…
arxiv.org25 days agoView details
arXiv:2608.20711v1 Announce Type: new Abstract: High-performance ML systems increasingly rely on GPU kernels whose editable source is unavailable, generated, or too distant from final machine code to expose remaining optimizations. Existing LLM kernel optimizers and autotuners mainly operate on CUDA, Triton, HIP, or t…
arxiv.org25 days agoView details
PrimeAgentOrchestrator: Memory-Primed Agent Spawning for Personal AI Infrastructure
arXiv:2608.20342v1 Announce Type: new Abstract: Large language model (LLM) coding agents start each session with an empty context window, discarding accumulated knowledge from prior work. We present PrimeAgentOrchestrator (PAO), a system that spawns new instances of Claude Code -- Anthropic's terminal-based coding age…
arxiv.org25 days agoView details
Truth Lies Deep: Countering Semantic Camouflage via Latent Intent Verification
arXiv:2608.20378v1 Announce Type: new Abstract: Safety alignment in Large Language Models (LLMs) is often superficial, relying on refusal mechanisms that trigger only at the final stages of generation without erasing the foundational knowledge of harmful concepts acquired during pretraining. This study demonstrates th…
arxiv.org25 days agoView details
arXiv:2608.20661v1 Announce Type: new Abstract: Enterprise adoption of large language models in finance is constrained less by fluency than by trust: in Financial Planning and Analysis (FP&A) and other regulated workflows, an answer is usable only if it is traceable to authoritative sources and auditable after the fac…
arxiv.org25 days agoView details
Why2Speak: Faithful Reasoning for Abstaining Action Policies
arXiv:2608.20670v1 Announce Type: new Abstract: Many agentic systems must repeatedly choose between acting and abstaining, making faithful reasoning important for oversight: an explanation is useful only if it reflects the computation that produced the action. We study this problem through intervention timing in multi…
arxiv.org25 days agoView details
CDRL: Certification-Driven Reinforcement Learning for Neutrino Flavor Model Discovery
arXiv:2608.20686v1 Announce Type: new Abstract: Many scientific discovery problems require searching combinatorial hypothesis spaces under complex domain constraints. Reinforcement learning (RL) offers a promising approach, but existing methods rely on scalar rewards that provide limited information about why candidat…
arxiv.org25 days agoView details
VortexChat: An agentic framework for autonomous multi-objective integrated photonic design
arXiv:2608.20688v1 Announce Type: new Abstract: The advancement of modern integrated photonics is frequently bottlenecked by device design workflows that rely heavily on manual simulation and expert intuition. While inverse design offers an alternative, it remains constrained by expert supervision and a lack of end-to…
arxiv.org25 days agoView details
Consilience: Conformally Calibrated Communication Control for Hidden-Profile Multi-Agent Reasoning
arXiv:2608.20564v1 Announce Type: new Abstract: Multi-agent LLM systems can improve reasoning by pooling diverse perspectives, but their effectiveness depends on coordinating communication, particularly in hidden-profile settings where each agent holds only part of the evidence required for a correct decision. Existin…
arxiv.org25 days agoView details
Volumetric Radiology AI in the Era of Multimodal Large Language Models
arXiv:2608.20549v1 Announce Type: new Abstract: Advances in multimodal large language models (MLLMs) are extending radiological artificial intelligence (AI) beyond task-specific image analysis toward multimodal understanding and reasoning. Volumetric radiology, however, presents a fundamental representational mismatch…
arxiv.org25 days agoView details
ExpertIVS: Sociological Expert Driven Individual Value Simulation in Large Language Models
arXiv:2608.20355v1 Announce Type: new Abstract: Large Language Model (LLM) agents have demonstrated considerable potential for social simulation, yet struggle to accurately model individual value systems. Most existing methods mechanically stitch survey responses into prompts, which suffer from semantic fragmentation,…
arxiv.org25 days agoView details
STCO: Conditional Neural Operators for Time-Dependent PDEs
arXiv:2608.20477v1 Announce Type: new Abstract: Neural operators have emerged as efficient surrogates for time-dependent physical systems governed by partial differential equations (PDEs), but their future-state predictions are often conditioned only on observed states and static problem descriptors. For control or op…
arxiv.org25 days agoView details
arXiv:2608.20649v1 Announce Type: new Abstract: Designers and policymakers in sociotechnical domains like content moderation, privacy interfaces, recommender systems and beyond, must choose among a growing menu of proposed interventions, but typically lack a principled basis for comparing them. Prior work tends to eva…
arxiv.org25 days agoView details
arXiv:2608.20388v1 Announce Type: new Abstract: Microservice placement in the compute continuum is driven by low-level Service-level Objectives (SLOs), but requiring users to specify metric-level constraints creates an adoption barrier and increases misconfiguration risk. Although large language models (LLMs) can inte…
arxiv.org25 days agoView details
Dual-Cache Latent Space Communication between Heterogeneous Language Models
arXiv:2608.20617v1 Announce Type: new Abstract: Multi-agent LLM systems split work across models, so answering often requires knowledge that sits in another agent's context: a Sharer has encoded information that a Receiver needs to complete its task. They usually communicate by exchanging text, which puts autoregressi…
arxiv.org25 days agoView details
arXiv:2608.20786v1 Announce Type: new Abstract: Multi-agent pipelines that author formal documents must both read a requester's forms and write against them. We report a deployed tender-response system, running an open-weights model under sovereignty constraints, and evaluate it against human-written bids the same org…
arxiv.org25 days agoView details
arXiv:2608.20757v1 Announce Type: new Abstract: We describe the PSK submission to the WMT 2026 Multilingual Instruction Shared Task. Our system uses the 3.35B-parameter Tiny Aya Global model with three QLoRA adapters, one for each task. The adapters are trained on multilingual document-summary pairs, passage-based que…
arxiv.org25 days agoView details
A Survey on Foundations and Frontiers of Multimodal Agentic Frameworks: Techniques and Applications
arXiv:2608.20379v1 Announce Type: new Abstract: Advances in large language models (LLMs) have fueled a wave of research into agency: the ability to reason, plan, and act. This effort has produced agentic frameworks that orchestrate perception, memory, and decision-making around powerful LLM backbones. With the advent…
arxiv.org25 days agoView details
Beyond Raw Transcripts: Structured Persona Extraction for LLM-Based Digital Twins
arXiv:2608.20344v1 Announce Type: new Abstract: LLM-based "digital twins" aim to simulate how an individual would behavein new environments or respond to novel questions, given some representation of that individual's prior responses. A common approach constructs this representation from survey transcripts or summarie…
arxiv.org25 days agoView details
FL-MAESTRO: Multi-Agent LLM Orchestration for Resource-Constrained Federated Learning
arXiv:2608.20518v1 Announce Type: new Abstract: In Federated Learning (FL), the communication topology is a runtime variable rather than a fixed design choice, since links and edge devices drop in and out during training. Each round, the server must commit three coupled decisions, namely the communication topology, pe…
arxiv.org25 days agoView details
arXiv:2608.20351v1 Announce Type: new Abstract: We ask whether stereotype-loaded queries about culturally marked people leak more personal information from a retrieval-augmented generation (RAG) system than otherwise-equivalent neutral queries. We pre-register a four-culture audit (en-Anglo, es-LATAM, Arabic, Hindi) o…
arxiv.org25 days agoView details
Environmental Slow AI: Design Principles for Generative Systems
arXiv:2608.20398v1 Announce Type: new Abstract: Generative AI (genAI) systems produce cultural artefacts at scale, but they also reflect embedded cultural values through their design. Once identified, these values become open to deliberate reshaping. This position paper examines the maximalist values of current genera…
arxiv.org25 days agoView details
TH-GNN: Heterogeneous Temporal Graph Neural Networks for LLM-Agent Shilling Attack Detection
arXiv:2608.20376v1 Announce Type: new Abstract: LLM agents can now generate realistic shilling profiles, fluent reviews, and coherent ratings at scale, systematically defeating recommender-system defenses. Text-only detectors that flag semantic drift in review embeddings are blind to graph structure and temporal coord…
arxiv.org25 days agoView details
Research Paper Quality Recognition Through Textual Feature Analysis
arXiv:2608.20368v1 Announce Type: new Abstract: Knowledge and innovations are shaped by using the quality and credibility of the scientific research. Yet, distinguishing between impactful, high-quality work and flawed studies remains a challenge. This paper introduces a benchmark for classifying research papers into t…
arxiv.org25 days agoView details
Evaluation-as-Search: Adaptive Discovery of Grounding Failures in Meeting Assistants
arXiv:2608.20392v1 Announce Type: new Abstract: LLM-powered meeting assistants are deployed at scale, yet systematic evaluation of their grounding fidelity remains limited to static benchmarks that miss failure modes tied to specific discourse structures or reasoning demands. We propose Evaluation-as-Search (EaS), a f…
arxiv.org25 days agoView details
No Judgment Without a Reason: Counterfactual Receipts for Versioned AI Evaluators
arXiv:2608.20938v1 Announce Type: new Abstract: Evaluators often produce correct labels via flawed reasoning, a critical failure for agentic systems gating actions, routing reviews, or supplying training feedback. Standard evaluation only verifies final label correctness, ignoring whether judgment changes stem from va…
arxiv.org25 days agoView details
Evaluating Skills, Not Just Agents: Agentic Continuous Evaluation of Skills
arXiv:2608.20614v1 Announce Type: new Abstract: Enterprise agent programs are moving from prototypes into production, where reusable skills, tools, and workflow packages must be reviewed with evidence rather than prose. Current gates often scan these artifacts for structure, style, and security, but they do not answer…
arxiv.org25 days agoView details
Terminal Agents: A Survey of AI Agents in Command-Line Environments
arXiv:2608.20485v1 Announce Type: new Abstract: Large language model agents increasingly act through terminals, yet existing surveys disperse terminal-mediated behavior across software engineering, tool use, and computer-use research. We regard terminal agents as systems whose dominant progress-bearing action--observa…
arxiv.org25 days agoView details
Graph-Operator World Models for Morphology-Parameter Generalization in Continuous Control
arXiv:2608.20936v1 Announce Type: new Abstract: World models for continuous control are commonly trained for a fixed physical system and can degrade when known morphology parameters such as link lengths, masses, damping, and actuation change. Existing approaches often provide these parameters as conditioning informati…
arxiv.org25 days agoView details
arXiv:2608.20390v1 Announce Type: new Abstract: General-purpose large language models (LLMs) are increasingly used to answer religious questions, but for Islamic content they carry two serious risks: factual fabrication (inventing Qur'anic verses or hadith) and subtle value misalignment. We present Ansari, a deployed,…
arxiv.org25 days agoView details
ImmigrationReason: A Structured Dataset of U.S. Immigration Appeals for Legal Reasoning Research
arXiv:2608.20391v1 Announce Type: new Abstract: Most legal NLP resources draw from federal case law and focus on coarse classification, leaving administrative adjudication, where the vast majority of government decisions occur, essentially unaddressed. We introduce ImmigrationReason, a large-scale structured dataset d…
arxiv.org25 days agoView details
CAS: Conformalized Agentic Search via Adaptive Retrieval and Policy Weighting
arXiv:2608.20771v1 Announce Type: new Abstract: Search Agents face a severe reliability crisis during reinforcement learning (RL) fine-tuning. Heuristic Top-K retrieval often causes critical evidence loss or noise inclusion, while over-confidence induced by progressive RL leads to hallucinated answers and redundant se…
arxiv.org25 days agoView details
arXiv:2608.20777v1 Announce Type: new Abstract: As scientific literature grows and papers increasingly under-report limitations, multi-agent LLMs offer a promising approach to systematically uncover these hidden failure modes. Here, we introduce Tree-of-Concerns, a multi-agent framework that deploys specialized skepti…
arxiv.org25 days agoView details
SDAD: Spec-Driven Agentic Development for the AI-Native SDLC
arXiv:2608.20341v1 Announce Type: new Abstract: Frontier coding agents backed by large language models with context windows from hundreds of thousands to millions of tokens are restructuring the Software Development Life Cycle (SDLC). Rich context handling and multi-step reasoning now allow substantial Functional Requ…
arxiv.org25 days agoView details
UpgradeBench: A Decision-Centric Benchmark for Upgrading Fine-Tuned LLM Specialists
arXiv:2608.20918v1 Announce Type: new Abstract: Organizations maintain task-specific adapters for open-weight language models, and each new base-model release forces a migration decision: retain existing specialists, port adapters, refresh from preserved behavior, or retrain. Prior transfer work evaluates isolated mod…
arxiv.org25 days agoView details
Difficulty-Aware Semantic-ID Optimization for Generative Recommendation
arXiv:2608.20611v1 Announce Type: new Abstract: Semantic-ID-based generative recommendation casts retrieval and ranking as autoregressive generation over hierarchical item identifiers. A common recipe is SFT followed by GRPO, yet vanilla GRPO is poorly matched to this tree-structured task. Under the frozen SFT checkpo…
arxiv.org25 days agoView details
Applying Anthropic Primitives at Large Enterprises: Harness Paradigm for Knowledge Work
arXiv:2608.20622v1 Announce Type: new Abstract: Frontier models have collapsed the cost of writing custom code: a niche problem a specialist sees in their own domain now costs an afternoon. The cost of reviewing and maintaining that code hasn't collapsed. Each solution drifts from the next; understanding one means rea…
arxiv.org25 days agoView details
arXiv:2608.20345v1 Announce Type: new Abstract: Conversational AI systems have become informal mental health support resources for Generation Alpha (Gen Alpha, born 2010-2024), with 13.1% of U.S. adolescents (5.4 million) using generative AI for mental health advice. While these systems, from therapy apps to general c…
arxiv.org25 days agoView details
arXiv:2608.20797v1 Announce Type: new Abstract: Evaluating language-guided mobile agents has recently shifted from rule-based to model-based approaches to achieve scalable and automated assessments. However, existing holistic evaluation paradigms process entire trajectories at once, leading to substantial context over…
arxiv.org25 days agoView details
arXiv:2608.20735v1 Announce Type: new Abstract: Manipulating moving objects requires a policy to anticipate contact events, yet vision-language-action (VLA) policies are commonly fine-tuned from the current observation alone. World action models (WAMs) learn predictive dynamics, but running a video-scale teacher or ex…
arxiv.org25 days agoView details
Who Delegates to AI? Evidence from 53,000 Agent Configurations
arXiv:2608.20425v1 Announce Type: new Abstract: A growing literature measures how far occupations are exposed to AI, but these measures capture where AI could perform tasks, not whether workers have adopted it. We propose a new layer of exposure, delegated exposure, which records whether a worker has committed a task…
arxiv.org25 days agoView details
Beyond Endpoint Gains: A Weight-Delta Audit of Medical Specialization
arXiv:2608.20768v1 Announce Type: new Abstract: Specialist language models are usually understood through endpoint gains: the generalist scores lower, the specialist scores higher, and the difference is treated as evidence of specialization. This leaves the released update itself largely unexamined. We propose a paire…
arxiv.org25 days agoView details
Quantization-Aware Healing: A Practical Recipe for Recovering Compressed, 4-Bit LLMs
arXiv:2608.20953v1 Announce Type: new Abstract: Serving large language models cheaply increasingly means shipping models that are both structurally compressed to a fraction of their parameters and quantized to 4 bits. Together these steps degrade reasoning, mathematics, coding, and long-context behavior enough to requ…
arxiv.org25 days agoView details
Dynamic Context Scheduling: Learning Beyond the Static Universe
arXiv:2608.20799v1 Announce Type: new Abstract: We study dynamic context scheduling as a training instrument for contextual re- inforcement learning. Rather than treating intra-episode context variation as a deployment reality, we treat it as a controlled shaping mechanism. Thereby, context evolves within each trainin…
arxiv.org25 days agoView details
Models
View all models →video-to-video · safetensors · video-generation · multiview-video-generation
huggingface.co25 days ago81 ptsView details
text-generation · transformers · safetensors · qwen3_5_moe
huggingface.co26 days ago80 ptsView details
text-generation · transformers · safetensors · ael
huggingface.co26 days agoView details
Open source
View all open source →<details open> vendor : update subprocess.h (#27409) </details> **Website:** - <https://llama.app> **Attestations:** - <https://github.com/ggml-org/llama.cpp/attestations/42402532> **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/down…
github.com26 days agoView details
<details open> cuda : add POOL_1D support (#27573) * cuda : add POOL_1D support * fix: add missing trailing newline for editorconfig compliance </details> **Website:** - <https://llama.app> **Attestations:** - <https://github.com/ggml-org/llama.cpp/attestations/42401257> **macOS…
github.com26 days agoView details