arXiv AI
Aug 11
On the Robustness of LLMs' Internal Representation of Code Correctness
arXiv:2608. 08266v1 Announce Type: cross Abstract: Code generated by modern language models often reads naturally.
Large language models
Model releases, architecture work and prompting research on large language models — from frontier-lab announcements to the arXiv papers behind them.
arXiv AI
Aug 11
Bidirectional Context Self-Distillation for Reinforcement Learning of Skill-Based LLM Agents
arXiv:2608. 09555v1 Announce Type: new Abstract: External natural-language skills provide large language model (LLM) agents with reusable and editable guidance for solving complex tasks.
arXiv AI
Aug 11
Matryoshka Language Model Suites
arXiv:2608. 09703v1 Announce Type: new Abstract: Training a language model suite classically requires training each model separately and serving them independently.
arXiv AI
Aug 11
Multi-Branch Policy Optimization for Multimodal Large Language Models
arXiv:2608. 07581v1 Announce Type: cross Abstract: Group-based reinforcement learning methods for multimodal large language models typically rely on trajectory-level credit assignment that applies a single advantage to all tokens in a response.
arXiv AI
Aug 11
Shape Mutating Expert Compression:LorExperts and BTExperts
arXiv:2608. 07814v1 Announce Type: cross Abstract: Mixture-of-Experts (MoE) language models deliver high capacity at low per-token compute, but deploying them cheaply requires compressing their many expert weight matrices.
arXiv AI
Aug 11
Spectral Outliers Reveal Dominant Learned Structure in Transformer Attention
arXiv:2608. 07921v1 Announce Type: cross Abstract: We apply Marchenko-Pastur (MP) random matrix theory to pre-trained attention weights in order to separate each projection matrix into a random-like bulk and a set of spectral outliers.
arXiv AI
Aug 11
Quokka: Accelerating Program Verification with LLMs via Invariant Synthesis
arXiv:2509. 21629v4 Announce Type: replace-cross Abstract: Program verification relies on loop invariants, yet automatically discovering strong invariants remains a long-standing challenge.
arXiv AI
Aug 11
SkillSentry: Reliable Skill Execution for LLM Agents via Runtime Assurance
arXiv:2608. 09253v1 Announce Type: new Abstract: LLM agents are increasingly equipped with skills to perform complex tasks through multi-step reasoning and tool use.
arXiv AI
Aug 11
Capability Is Not Propensity: Measuring Pressure-Robust Cooperative Behavior in Civic LLM Agents
arXiv:2608. 09485v1 Announce Type: new Abstract: Cooperative capabilities in language models are dual-use.
arXiv Machine Learning
Aug 11
RA-FinBERT: Rule-aware LoRA adaptation for low-resource financial sentiment classification
arXiv:2608. 09834v1 Announce Type: cross Abstract: Financial sentiment analysis converts unstructured financial news into quantitative signals that can support market analysis and decision-making.
arXiv Machine Learning
Aug 11
Can Webcam Gaze Constrain Mesa-Objectives in Driving Models? An Instrument Precision Analysis
arXiv:2608. 08947v1 Announce Type: cross Abstract: Current hazard detection systems in autonomous driving may develop mesa objectives, learned internal goals that achieve high training performance through spurious correlations rather than genuine hazard recognition.
arXiv AI
Aug 11
Directed Neuro-Symbolic Stochastic Execution for Verification of Distributed Parallel AI Programs
arXiv:2608. 07947v1 Announce Type: new Abstract: Distributed parallel Artificial Intelligence (AI) programs expose reliability gaps that conventional testing cannot close: parallel executions are non-deterministic, and AI workloads bring high-dimensional inputs and non-linear operations that defeat fuzzing and symbolic execution in isolation.
arXiv AI
Aug 11
An evolutionary model of animats with VLM-based subjective evaluation
arXiv:2608. 07537v1 Announce Type: cross Abstract: In this study, we propose a framework that incorporates subjective evaluations provided by a Vision-Language Model (VLM) into the fitness evaluation and selection processes of a genetic algorithm.
arXiv AI
Aug 11
Evolving Safety Landscape of Multi-modal Large Language Models: A Survey of Emerging Threats and Safeguards
arXiv:2608. 07535v1 Announce Type: cross Abstract: Multi-modal large language models (MLLMs) integrate heterogeneous modalities through modality alignment and fusion, enabling stronger understanding and reasoning.
arXiv AI
Aug 11
BRACE: Taming Sharp Irregularities via Barycentric Rational Forecasting for Fast Diffusion Transformers Inference
arXiv:2608. 07572v1 Announce Type: cross Abstract: Diffusion Transformers (DiTs) have demonstrated exceptional performance in high-fidelity image and video generation.
arXiv AI
Aug 11
Can LLM Agents Stick to the Script? A Benchmark for Long-Horizon Consistency in Interactive Narratives
arXiv:2608. 08160v1 Announce Type: cross Abstract: The rapid advancement of Large Language Models (LLMs) is revolutionizing AI for Games by enabling open-ended and fluid interactive storytelling.
arXiv AI
Aug 11
SkillConsist: Detecting Inconsistencies in Agent Skills via Bidirectional Graph Alignment
arXiv:2608. 07639v1 Announce Type: cross Abstract: Agent Skills provide reusable capabilities to LLM agents.
arXiv AI
Aug 11
PROSLEX: A Novel Dataset for Expert-Annotated Legal Statute Prediction for Indian Judiciary
arXiv:2608. 08830v1 Announce Type: new Abstract: Legal Statute Prediction (LSP) involves automatically identifying relevant legal statutes given factual descriptions in legal documents, typically framed as a multi-label classification task within natural language processing and information retrieval research.
arXiv AI
Aug 11
Findings of the First Teaching Monster Challenge: A Benchmark of Pedagogical Content Knowledge in AI Agents
arXiv:2608. 08852v1 Announce Type: new Abstract: AI agents can now solve problems, answer like subject experts, and generate long-form multimodal content.
arXiv AI
Aug 11
Full-bandwidth transformer
arXiv:2608. 08888v1 Announce Type: new Abstract: Autoregressive transformers compute along two axes: horizontally across generated tokens, and vertically through model depth.