GAIA

Headlines · 4

1. Oct 22
   HybridAgent paper mixes latent-space thinking with text tool calls

   A commentary published by Huxiu AI reports that a paper submitted to arXiv on October 13 (2610.18861v1), with authors drawn from Stanford, UIUC, Together AI and Meta, proposes HybridAgent, which lets a model decide within a single generation whether to reason in latent space without emitting text or to act in text for tool calls and outward delivery. The design rests on three parts: HybridLink, a residual adapter of roughly 10M parameters on a 4B model that switches between text and latent modes; a continuous working memory that carries mixed context across turns; and a pseudo-text rendering of latent reasoning that monitors can read. The authors self-report 2.7–3.7x cost savings, 1.8–2.6x speedup and a 14B model beating GPT-5.6-Luna by about 14.5 points on the single-point GAIA benchmark, and the article states none of these figures has been independently replicated.

   虎嗅 AI · 🔥 0
2. Oct 19
   Self-Credit paper adds dense credit assignment to long-horizon agentic RL

   A new arXiv paper (2610.28388, submitted 20 Oct 2026) by Fangcong Yin, Ming Yin and colleagues proposes Self-Credit, an end-to-end agentic reinforcement learning method that turns sparse outcome rewards into dense supervision signals via self-generated credit assignment and intra-group rescaling. The method builds on group-based RL with three parts: the policy model writes summaries citing earlier turns that held critical evidence, intra-group rescaling compares cited actions across successful and failed rollouts of the same prompt, and advantage reweighting applies the rescaled weights to the cited turns. The authors report that Self-Credit improves over base models by 37.0 and 36.2 points on BrowseComp-Plus for Qwen3 8B and 32B, beats the strongest agentic RL baselines by 8.5 and 5.6 points there, and leads the strongest baselines by 4.9 points on GAIA and 4.7 points on xBench-DeepSearch. These figures are the authors' own and have not been independently replicated.

   Hugging Face · Papers · 🔥 0
3. Oct 9
   Wayve and Stellantis detail AI Driver rollout in Fiat, Jeep and Maserati, targeting 2028

   At the Wave by Vento event in Turin, Wayve CEO Alex Kendall and Stellantis CEO Antonio Filosa detailed their partnership: the Wayve AI Driver will run on Stellantis's STLA AutoDrive platform for hands-free, supervised driving, with a first launch targeted for North America in 2028. Demo rides were given in a Fiat 500e and a Maserati Grecale, and Kendall said the system is tuned per brand and should run on under $1,000 of hardware. Kendall framed supervised driving as a step toward eyes-off and driverless systems, and noted Wayve also has a Mercedes-Benz production deal and a supervised robotaxi service with Uber in London.

   The Next Web · 🔥 0
4. Oct 8
   Anthropic commits $150M to Genesis Mission; Claude Science maps sky in UV

   Anthropic said on Oct. 8 that it will commit $150 million over three years to the US federal Genesis Mission, making Claude available to more than 15 participating agencies including NASA, the National Institutes of Health and the National Science Foundation, and providing Claude, Claude Code and API credits to several hundred research projects. The company announced the pledge at the White House Office of Science and Technology Policy's Science: A New Golden Age summit in Washington. In a separate result described by Anthropic and covered by outlets including The Decoder and cnBeta, Johns Hopkins astrophysicist Brice Ménard used Claude Science to combine public ultraviolet data from multiple space missions and fill gaps with statistical inpainting, producing what Anthropic calls the first complete UV map of the sky; Anthropic says predictions averaged about 10% deviation from measurements and that work it says would have taken weeks took days, while a commenter objected that inferred areas are not real observations and should not be called complete.

   Anthropic · News · 🔥 0

Experience and discussion from the community

Share my GAIA experienceAsk about GAIA