A new framework from arXiv (2608.14613v1) introduces a mechanism-design approach to verify and repair LLM-agent negotiations over Anthropic's Model Context Protocol (MCP) and Google's Agent2Agent (A2A…
A new arXiv paper (2608.14711v1) reveals that AI coding agent benchmarks misapply the pass@k estimator, inflating reported scores by 0.85-0.97 in absolute terms (0.96-0.98 reported vs. 0.00-0.12 corre…
A new arXiv paper (arXiv:2608.14791v1) introduces CEDAR-GRPO, a process-aware reinforcement learning framework that improves abductive reasoning in large language models by combining final-answer corr…
A training-free framework called ACTS-SQL, developed by researchers and deployed in Volcano Engine's Torch Log Service (TLS), improves SQL execution accuracy from 36.77% to 53.61% on real user queries…
Researchers propose Adversarial Translation Optimization (ATO), a gradient-based method that augments existing machine translation benchmarks to increase difficulty, lowering average translation quali…
A new benchmark called MecEng, introduced in an arXiv paper (2608.14615v1), evaluates large language models on mechanical engineering tasks, finding that the best open-weight model achieves an 86.0% s…
Researchers introduced Pistis, an agentic proof search tool that generates faithful formal Lean proofs, and applied it to Euclid's Elements, producing proofs favored 2.89x by human reviewers and 5.2x …
A new arXiv paper (2608.15304v1) systematically analyzes risks induced by expanding cognitive capabilities in frontier agentic systems powered by large language models (LLMs), proposing a three-level …
CacheScout, a new KV-cache runtime layer for multi-agent LLM serving developed by researchers, improves KV-cache hit rate by 10-18 percentage points, reduces mean time-to-first-token (TTFT) by 18-45%,…
A new arXiv study (2608.15693v1) finds that no single model compression technique wins across edge AI tasks, reporting that Qwen3.5 0.8B reaches 93.85 SQuAD F1 and 92 EM under Q5_K_M GGUF quantization…
A benchmarking study of 14 large language models (LLMs) found that they are not reliable sources of cosmetic chemistry information, with overall performance poor and most pronounced deficits in quanti…
Researchers introduced GraphLoom, a reliability-calibrated multimodal knowledge-graph retrieval-augmented generation framework that improves answer quality and evidence faithfulness by routing compact…
A new study from arXiv (2608.14651v1) evaluating open-weight Multi-Modal Large Language Models (MM-LLMs) for disaster assistance found that no model achieves reliable consistency across text and audio…
A new arXiv preprint (arXiv:2608.15382v1) proposes a causal knowledge graph-based evaluation framework for healthcare large language models (LLMs), tested in a cardiovascular pilot. The framework's in…
Researchers introduced ESTP (Entropy-and-Semantic Token Pooling), a lightweight framework that combines entropy with attention-based importance scores to improve LLM output length prediction, outperfo…
A new study applies machine learning and natural language processing to aviation safety data from Socrata, the Australian Transport Safety Bureau, the National Transportation Safety Board, and the Avi…
A new arXiv study (2608.14992v1) found that tool-result records carry more authority than plain text in influencing Claude Opus 5's answers, but the effect was not consistent across replications. In t…
Nigeria-based defense technology startup Terra Industries has raised $52 million in what it says is the largest-ever seed funding round for an African defense tech company. The company develops automa…
Cursor introduced Origin, a cloud code hosting service, marking its first major product update since its $60 billion sale to SpaceX Corp. in June. The service competes directly with GitHub, leveraging…
CNBC's Jim Cramer said the AI boom has transformed the historically cyclical memory industry, with demand and greater supply discipline supporting stronger profits, and he identified four soaring memo…