SpecBench: Measuring Reward Hacking in Long-Horizon Coding Agents
Explorar
Noticias de IA
37834 elementos — filtrados, clasificados y sin duplicados
OpenDiscoveryTrace: Process Traces for Evaluating AI Scientist Workflows
Black-Box Red Teaming of Agentic AI: A Taxonomy-Driven Framework for Automated Risk Disco…
Complementing reinforcement learning with SFT through logit averaging in the post trainin…
Improving Cross-Lingual Token Representations by Adding a Pinch of SALT
Compact Visuotactile World Models for Lifting: Prediction, Reward Alignment, and Force Co…
RAU: Reference-based Anatomical Understanding with Vision Language Models
When Auditors Fabricate: Batch-Size Degradation and Confident Hallucination in LLM Detect…
An Autonomous GeoAI Agent for Arctic Eco-Navigation
Introducing Consort: A Spec-First Agent Framework for Enforced, Test-Driven Development o…
Looped GPT-BERT: Trading Parameters for Computation in Small Language Modeling
From Plausible to Actionable: A Position on LLM Self-Explanations
ConvMem: Convolutional Memory for Long-Context Reasoning
A Taxonomy of Architecture Options for Foundation Model-based Agents: Analysis and Decisi…
CT-SAFR: Safe and Interpretable Chain-of-Thought Reasoning for Autonomous Robots: A Multi…
The Era by Eon Benchmark: A Generated Enterprise Estate with Exact Ground Truth for Bench…
Predicting Estimated Times of Restoration for Electrical Outages Using Longitudinal Tabul…
EVOQUANT: Self-Evolving Verifier-Guided Strategy Optimization for Robust Quantitative Tra…
Which Tokens Should SFT Actually Learn? A Token-Trimming Perspective on Mathematical Reas…
AMD-Backed Korean AI Startup Upstage Considers Pre-IPO Round
Alibaba Backs Ex-Staffer’s AI Testing Lab at $2.5 Billion Value
Stonepeak CEO on Data Centre Investment Outlook
Shield AI on Business Plans
DOJ Probes Nvidia’s License Deal With Groq on Antitrust Concerns
With the National Football League (NFL) back tonight, love seeing Seattle Seahawks analys…
Training a 3.8B LLM to 0.384 CORE for $998 – Hugo Vergnes
A Stupid Idea for AI Alignment We Came with by Looking at Specification Gaming
AI research startup Listen Labs scrubbed a $1.5B funding round for Salesforce talks
Rebuilding AUTOMATIC1111 with Gradio Workflow
Build more natural voice experiences with GPT‑Live‑1 in the API