Research / AI/LLM Reasoning
arXiv paper on distilling multi-agent debate into a single LLM with a two-stage fine-tuning pipeline. The abstract reports lower token use, comparable or better benchmark performance, and an analysis of agent-specific activation subspaces, with code linked from the page.
The page clearly describes an AI research contribution focused on LLM reasoning and post-training methods, with enough abstract detail to understand the approach, claimed results, and practical implications. It also points to code, which makes it relevant for AI builders and researchers.
The crawl does not show the code repository itself, so visitors cannot tell the implementation status, licensing, setup steps, or whether the code is currently usable beyond the paper link.
Last evaluated 15 days ago. Current rank #666. Down 1 spot in the rankings.
An interactive dashboard that analyzes New York Times coverage since 2000 using the NYT Archive API, with views for reporters, beats, sections, subjects, geography, obituaries, and corrections.
An open benchmark and leaderboard for AI CAD agents, with 308 prompts across 20 categories and layered scoring for geometry, engineering, manufacturability, and cognition.
A research article from Applied Compute on how agentic, tool-using workloads differ from traditional LLM benchmarks, with production observations, workload profiles, and an open-source harness for replaying traces.
A research code repository for studying how fine-tuning can trigger verbatim recall of copyrighted books in large language models. It includes preprocessing, fine-tuning, generation, and memorization-evaluation scripts, with setup notes and example data.
