Agents & tool use Papers

A 2020 paper that pairs a pre-trained seq2seq model with a dense Wikipedia index fetched by a neural retriever, setting state of the art on three open-domain QA tasks and producing more factual text than a parametric-only baseline.

ReAct is a prompting method that has a language model interleave written reasoning with calls to outside tools or environments, improving question answering, fact verification and interactive decision-making.

A benchmark of 2,294 real software engineering problems taken from GitHub issues and the pull requests that resolved them, in which a model must edit a codebase to fix an issue; on release the best model solved under 2% of them.

Kimi K2 is Moonshot AI's trillion-parameter mixture-of-experts model, pre-trained with the MuonClip optimizer and post-trained on large-scale agentic data plus reinforcement learning; the report reports state-of-the-art results among open-source non-thinking models and releases base and post-trained checkpoints.

Zhipu's GLM-5 report describes a 744-billion-parameter mixture-of-experts model built on sparse attention and asynchronous reinforcement learning, aimed at moving AI-assisted programming from vibe coding toward agentic engineering.

A 2023 paper by Schick, Dwivedi-Yu, Dessì and co-authors showing that a language model can teach itself when to call a calculator, search engine, Q&A system, translation system or calendar, and how to use the results.

WebArena is a reproducible environment of four fully functional websites — e-commerce, a social forum, collaborative software development and content management — built for testing language-guided web agents. Its best GPT-4 agent completed 14.41% of tasks; humans completed 78.24%.[^1]

GAIA is a benchmark of real-world questions that are easy for humans but hard for AI assistants, testing reasoning, multimodality, web browsing, and tool use. It compares AI progress against average human performance.

OSWorld is a benchmark and environment for testing multimodal agents on 369 real computer tasks in Ubuntu, Windows and macOS; humans complete over 72.36% of the tasks, while the best model reaches only 12.24%.

τ-bench is a benchmark, submitted to arXiv in June 2024, that tests language agents in dynamic conversations with a simulated user, with domain-specific API tools and policy rules, and scores them by database state rather than by the text of their replies.

Terminal-Bench 2.0 is a curated benchmark of 89 hard tasks carried out in computer terminal environments, built to measure how well AI agents handle realistic command-line work; frontier models and agents score below 65% on it.