Full-text search across 17145 articles. Combine with topic and date filters; results sorted by relevance.
Anthropic announced that its Claude AI agents autonomously discovered a novel enzyme system, which the company calls array-associated reverse transcriptases (ART), found in bacteriophages and associated with an array of …
Researchers submitted CliffCompaction, an autocompaction technique for long-horizon coding agents, to arXiv on 22 Sep 2026, reporting cost reductions of up to 50% under a bounded context while maintaining or improving pe…
Researchers submitted SpeakerMem-R1, a speaker-centered dual-track memory system for multi-party dialogue, to arXiv on 22 Sep 2026. SpeakerMem-R1 achieved 62.33% on the publicly reported EverMemBench leaderboard from Eve…
Researchers introduced SWE-Serve, a benchmark of 53 repository-grounded tasks derived from recent production changes to SGLang, spanning six inference engineering families, to evaluate agents on production inference engi…
A March 2, 2026 arXiv paper by Aditya Anirudh Jonnalagadda proposes the b-posit, a bounded posit format that caps the regime field at 6 bits and reports 79 percent lower power consumption, 71 percent smaller area, and 60…
An arXiv paper posted on 15 September 2026 presents an exposition of a proof of the Erdős–Sós Conjecture, which states that every graph with average degree greater than t-2 contains every tree on t≥2 vertices, attributin…
A new benchmark called FrontierMath Erdős (FME), introduced in a paper submitted to arXiv on 6 September 2026, evaluates AI systems on 68 Erdős problems that remained open as of August 2026, requiring them to prove or di…
A September 21, 2026 arXiv paper submitted to the Computer Science > Machine Learning category identifies "sequential reappearance," a failure mode in diffusion data-point unlearning where an instance initially judged fo…
A pre-registered arXiv study submitted 20 Sep 2026 found that prepending a 267-word specification frame to prompts cut defects in LLM-generated backend code across all five frontier models tested, with the frame arm winn…
Researchers Ying Wang and colleagues introduced temporal straightening, a curvature regularizer that encourages locally straightened latent trajectories in a Joint-Embedding Predictive Architecture (JEPA) world model, ac…
DeepSeek published a technical report on DeepSeek Elastic Compute (DSec), a production sandbox platform for large-scale agentic training and evaluation with large language models, submitted to arXiv on 19 Sep 2026. A sin…
A September 21, 2026 arXiv paper introduces Human-Agent Audit Collaboration (HAAC), a workflow and system for structuring human-AI collaboration in AI auditing, evaluated with 71 auditors on conversational shopping agent…
Researchers posted a paper on arXiv on 21 Sep 2026 introducing Burnashev Adaptive Posterior Matching (BAM), a feedback-coding scheme for black-box LLM steganography that recasts covert communication as a sequential probl…
A paper submitted to arXiv on 6 April 2026 shows that two AI agents operated by different entities can conduct an undetectable parallel secret conversation whose transcript is computationally indistinguishable from an ho…
Depop chief technology officer Keyur Govande, who joined the company in May 2025 after more than 14 years at Etsy, is focusing on AI-powered listing tools that automatically generate titles and descriptions from merchand…
Marist University and IBM announced a new Innovation Incubator on Sept. 22, 2026, anchored by an IBM z17 mainframe that will give Marist students and faculty on-campus AI computing power for research projects, expanding …
An eye-tracking study of 53 participants submitted to arXiv on 21 Sep 2026 found that how AI coding interfaces render generated code changes programmers' visual attention, with dynamic rendering producing fewer but longe…
A September 2026 comparison of open-source coding models found GLM-5.3-Flash from Z.ai best for agentic coding, DeepSeek V4 Flash cheapest per token, and MiniCPM5-2B best for on-device use. GLM-5.3-Flash, a 320B mixture-…
OpenBMB released VoxCPM2, a 2-billion-parameter, Apache-2.0 licensed text-to-speech model that clones voices from a few seconds of audio, generates new voices from text descriptions, and outputs 48 kHz speech in 30 langu…
A paper submitted to arXiv on 20 Sep 2026 introduces ArtifactBench, a lineage-aware evaluation suite for AI-generated music detectors that separates calibration from final testing and reports source-level performance wit…