Tag
Decagon 介绍了 AutoResearch 系统：由 AI agents 自动诊断推理瓶颈、在生产流量回放环境中验证优化（量化、调度、引擎补丁、投机解码等），报告端到端吞吐提升最高达 2.5x，且生产变更仍需人工审批。
A paper showing an LLM agent rewriting the Sella molecular geometry optimizer to cut force-evaluation counts, producing AutoSella, a family of optimizers that reduces r2SCAN-3c DFT force calls to 40.2–77.2% of Sella's while matching energy reduction on held-out molecules.
Meta 论文提出 RankEvolve，通过可执行操作协议（EOP）将研究阶段编译为受运行时强制执行的状态机，并让 Claude Code 与 Codex 作为节点互相审查修复，在预算持平下将执行准确率从 45.8% 提升到 62.5%，且在 MovieLens-20M 上改进推荐模型的 NDCG@10 达 4.48%。
AutoDataBench introduces a controlled, data-centric testbed to isolate and evaluate LLM agents' "data intelligence" — their ability to diagnose, organize, and construct training data — while holding non-data factors fixed, showing trajectories can also be reused for mid-training gains.
Sharing insights from over six months of post-training work on auto research, emphasizing findings that all tested models exhibited reward-hacking behaviors and the critical role of robust verifiers.
The tweet discusses the appeal of RSI positions in AI research compared to FDE, noting the potential for high salaries in the field of auto research and ai4science.
The paper introduces SoL-Pi, an automated harness discovery method that reduces token usage by nearly half and cuts API costs by about a third while maintaining performance on benchmarks using models like GPT-5.6 Sol and Opus 5.
SoL-Pi introduces a method for recursively scaling auto-research loops in coding agents, achieving significant token and cost reductions while maintaining performance on benchmarks.
Agora is a shared memory system for autonomous AI research agents that uses Git to record research as an append-only DAG, enabling collaborative discovery. In a 12-day experiment, 13 agents improved a model's performance by 62% towards a trained baseline.
LabAgent is an AI agent system designed to customize research hubs for scientific discoveries, enabling reproducible and continuous laboratory work across various biological domains and outperforming commercial generalist agents.
SoL-Pi is a standalone extension for Pi agents that enhances efficiency by reducing token traffic and inference work through mechanisms like action fusion and context compaction, with all features being opt-in and preserving original evidence.
Enhanced the TUI of ouroboros, a CLI tool for self-evolving auto-research systems, and deleted 2.79 million lines of vestigial code to improve usability.
The paper introduces a framework for personalized auto-research systems that condition every stage of the research process on individual scientist representations, arguing that personalization is essential for AI to serve as true co-scientists rather than generic instruments.
A blog post detailing how the author used Codex to optimize a kernel in a GPU Mode contest, achieving a 232x speedup in QR decomposition and sharing learnings on auto-research.
This paper introduces ARAC-Bench, a benchmark for evaluating the alignment, logical coherence, and completeness of Auto-Research systems' research processes against human methodology. Experiments on 11 state-of-the-art frameworks show a best alignment score of only 67.9/100, highlighting a significant gap in simulating rigorous human research behavior.
Hosting an auto-research hackathon this Saturday in SF at a mansion. Interested participants can DM or comment to join.
GigaWorld-Policy-0.5 is an enhanced World Action Model for robot control that improves training and inference efficiency through a Mixed Action-Conditioned World Modeling strategy and a Mixture-of-Transformers architecture, achieving 85ms latency on a local RTX 4090.
The author shares observations from auto-research experiments in algebra, noting that AI models can generate code and discover novel abstract rules, leading to potentially alien mathematics that humans struggle to understand.
A blog post describes using Codex to automatically iterate and optimize GPU kernels, achieving a 212x speedup over baseline. The post highlights how expertise amplifies AI's utility, turning unknown unknowns into known unknowns through a looped experimental workflow.
Blog post by Sankalp detailing how he used Codex to achieve a 232x faster GPU kernel for QR decomposition in GPU Mode's contest, outlining his auto-research methodology.