Excited to introduce 🎶 EvoDuet: co-evolving web search and task solutions to push the state of the art across 8 scientific optimization tasks.

🔬 Can AI discover unknown scientific knowledge by itself?

Excited to introduce Principia: Test-Time Hypotheses-to-Principle Construction for Scientific Discovery.

Principia uses LLMs to turn experimental observations into scientific principles at test time:

🔍 From ses
Compare related inputs and their outcomes to propose explanations for how specific characteristics contribute to the observed results.

🧠 From hypotheses to principles
Generalize recurring relationships across hypotheses into broader principles that guide subsequent predictions and candidate generation.

🔄 Refine knowledge with new observations
Use new observations to revisit earlier hypotheses and reconstruct the principles built on them.

Huge thanks to my co-authors at @KAIST_AI: @SangwooPark0306, @jinheonbaek, @SoyeongJeong97, and @SungJuHwang1

📄 Paper: arxiv.org/abs/2610.32195

#AIforScience #ScientificDiscovery #LLM #TestTimeLearning #KnowledgeConstruction

🚀 Terminal-Bench-Science is now featured in its first open-weight model release!

Kimi K3.1 Pro scores 38.1%, the strongest open-weight result on Terminal-Bench-Science to date.

Still, Terminal-Bench-Science shows BY FAR the biggest gap to SOTA (~30 points), a promising sign that it’s one of the most discriminative measures of frontier capability out there.

Great paper from Anthropic on AI agents doing ML-for-science research.

(bookmark it)

If you build AI-scientist systems, this one is worth your time.

They propose Agent4Sci, which gives frontier coding agents 20 scientific ML benchmarks across 11 domains and asks them to beat l.

They beat it on all 20. Claude Opus improves on the published best by 10.4% on average, and no run was flagged for misconduct.

The gains come almost entirely from standard ML-engineering tricks. The trajectories show agents misjudging what is novel, leaving compute unused, and settling into small local sweeps.

Adding three lightweight skills for research ideation, experiment management, and scheduled checklists raises Claude Opus from 10.4% to 20.1% and Codex from 4.4% to 8.1%, with more varied and more domain-relevant changes.

The models stay the same, and the research scaffolding roughly doubles the gain.

arxiv.org/abs/2610.28763

Chat with Paper: academy.dair.ai/papers/automat…

💡 (1/n) How do scientific ideas 𝐚𝐜𝐭𝐮𝐚𝐥𝐥𝐲 𝐞𝐯𝐨𝐥𝐯𝐞 from an initial question to a published discovery?

We introduce the 𝐑𝐞𝐬𝐞𝐚𝐫𝐜𝐡 𝐏𝐫𝐨𝐠𝐫𝐞𝐬𝐬 𝐆𝐫𝐚𝐩𝐡 (𝐑𝐏𝐆), a new representation to model 𝐚𝐮𝐭𝐡𝐞𝐧𝐭𝐢𝐜 𝐫𝐞𝐬𝐞𝐚𝐫𝐜𝐡 𝐭𝐫𝐚𝐣𝐞𝐜𝐭𝐨𝐫𝐢𝐞𝐬 It captures how ideas evolve through experiments, insights, failures, and revisions—and crucially, the reasoning behind every transition.

Existing approaches either generate research ideas in a one-shot manner or rely on artificial workflows that capture only fragments of the RPG.

📄 arxiv.org/abs/2610.27337

1/9 Current optimizers are static during training. What if the optimizer could think and react while the network trains?

We let an agent watch the training dynamics and rewrite the optimizer's source code as training goes: a reactive optimizer.

In every language and image as no tuned recipe, it matches or beats the best recipe one can one-shot, and it guards against loss spikes and overfitting by itself. In the near future, optimization decisions may belong to agents, not humans.

📝 xingjianbai.com/reactive-optim…
💻 Try it on your task: github.com/xingjian-bai/o…
🧵

🔥 Excited to share Agent World Model in the Wild (WildAWM)! 🌍

Fully open-source synthesis 🔓: No proprietary models needed! We synthesized 341 high-quality environment–task pairs, addressing 3 challenges in task synthesis:

✅ Solvability: At least one agent can successfully Difficulty: Challenging even for frontier models, with GPT-6-Astra achieving only ~72.7 pass@8.

🎯 Verifier reliability: 95% agreement between task verifiers and agent-as-a-judge ensemble.

📈 Effective RL training: Trained exclusively in these synthetic worlds, Qwen3.6-35B-A3B achieves an average +10.58-point improvement across 8 OOD benchmarks!

👋 KRAFTON's Raon is entering the open-weight scene!

Meet Raon-1 mini, our first model, trained from scratch.

On par with leading models in its size class, while running faster and using less memory.

Available under the Apache 2.0 license.
huggingface.co/KRAFTON/Raon-1…

🤩 Introducing Self-HPO: let the coding agent improve its own harness, then train a stronger policy model with the optimized harness, and repeat the process for recursive improvement!

💪 Self-evolving: no external model required
🎯 Harness search targets the policy's evolving r SWE-Bench Verified/Multilingual/Pro: 57.8% → 64.8%
🧵[1/n]

We looked at TerminalBench 4.0 and found a lot of issues. After fixing the issues Sonnet 5.5 scores 89.2%. So, TerminalBench is already mostly saturated.

Along the way we found a lot of crazy stuff, including reward hacks in the published transcripts.

🚀 Introducing DiscoveryRSI: Recursively Learning to Discover Valuable Hypotheses

6.3% of DiscoveryRSI-27B's hypotheses match later published work, vs 2.5% GPT-6 & 2.3% Claude Opus 5.

📄 arxiv.org/abs/2610.23839
💻 github.com/Gen-Verse/Disc…
🔴 foundry-arena.rsi-arena.workers.dev/arena

We are bringing agents to the wet-lab bench at Google Research to navigate complex biological landscapes. Learn how we are grounding agents in physical lab robotics to run fast closed-loop molecular design cycles, ultimately discovering single-digit nanomolar protein binders for /t.co/iPesR5MXCd

Autoresearch with coding agents can now run for days on a single problem, but we've seen an increasing number of cases where they stop improving early. We address this issue with a simple idea: forgetting.

Introducing our new paper: "Scaling Discovery Requires Forgetting"

Scientific 🔬 progress often starts with an unexpected observation that sparks a new hypothesis.

But AI agents often start with a hypothesis, then look for observations to support it.

We believe agents should actively explore both observations and hypotheses. When observations hypothesis testing, we risk missing discoveries we weren’t looking for.

⚡ Meet SHELDON: a scientific reasoning harness that models separate hypothesis and observation spaces and generalizes across medical diagnosis, system-log investigations, ARC-AGI 2, and even autoresearch.

Bazinga!

🧪Can AI agents take a scientific paper and write a meaningful follow-up to it?

We introduce Into the Unknown, an AI agent benchmark of 15 open scientific research problems. Each problem comes from a scientist currently or formerly at US DOE National Labs:
@LosAlamosNatLab b @AmesNatLab @PPPLab @Livermore_Lab

Can AI agents actually do research on problems nobody has solved yet? 🔭

In Into the Unknown, we give agents a real science paper + a research direction and ask them to write the follow-up paper.

Across 9 frontier agents and 15 open problems:
- None of the 135 submissions met all evaluation criteria.
- 114/135 failed integrity checks, often claiming work that was never done or reporting numbers that contradicted the recorded evidence.

Really fun to be part of this one - check out @seungonekim 's thread 👇

The Agentic Scientist: Orchestrating Coding Agents for Automated Scientific (Co-)Discovery

Blog: pub.sakana.ai/agentic-scient…
GitHub: github.com/SakanaAI/AI-Sc…
arXiv: arxiv.org/abs/2610.18933

Since 2024, we’ve used The AI Scientist as a way to test how far AI alone can take scientific to the paper published in Nature earlier this year. The Agentic Scientist is the next step in that line of work, built for the era of coding agents.

Coding agents such as Codex and Claude Code now carry out long software tasks on their own. Handing them a research problem, however, tends to produce a strong hyperparameter optimizer rather than a scientist: scores go up, but ideas stay narrow, and the reasons for the gain are rarely examined.

The Agentic Scientist makes the research loop explicit. It wraps off-the-shelf coding agents in a scaffold that controls what they see, what they write, how they are evaluated, and how results are handed to the next stage.

Parallel workers test hypotheses against rubrics defined before any experiment runs, share successes and failures through a common memory, and then ablate the winning program to generate evidence.

We used an early version of the system to study Error Diffusion, a biologically motivated alternative to backpropagation. Our researchers set the question and the rubric, chose directions at each stage, and verified the results, references, and claims before submission. The resulting paper was accepted as an oral presentation at ALIFE 2026. In autonomous runs, the system also raised Doc-to-LoRA retrieval accuracy from 52% to 98% on multi-needle tasks and designed a learned evolution strategy that improves over its CMA-ES reference on unseen tasks.

Alongside it we release AutoIllustrate, a pipeline for scientific figures that keeps an editable draw.io file as the source of truth, so every diagram can be corrected by hand.

Humans define the question and the standards of evidence; agents expand the experiments researchers can run. We call this Co-Discovery, and we are releasing both systems as open source for the community to build on.

What can large-scale AI models bring to the growing complexity of particle physics?

@CERN and Mistral join forces to explore this question.

The ambition: a single Large Physics Model, trained on billions of simulated collisions.

Your LLM agent has to think out loud, token by token, before it can search, run code, or act in the world.

We humans don’t. We ponder in silence and speak when it matters.

Human language was never meant to be the limit of machine thought.

Introducing 🌗Hybrid Reasoning Agents
, act through text
- Two modes, one unified multi-turn flow
- No black box, fully readable latent thoughts
- Any LLM can become a HybridAgent

AI is cracking math problems. What's holding it back in engineering?

The cost of evaluation. Agents can generate ideas in seconds, but testing them can take hours or days.

💡 Meet DuoDiscover: where evaluation is part of what agents discover 👇🧵

Some highlights:
🏆 #1 on 🐍 2× the experts on SWE-fficiency
🚀 1.92× on vLLM; 1.44× on FreeToken
🔥 2.68× over production TPU kernels
🖥️ Beats the best human and AI-designed CPU prefetchers

🔬 Calling scientists across disciplines: help build SciVeri-Bench v1.0!

After showcasing SciVeri-Bench v0.1 at the Frontier Data Summit hosted by @SnorkelAI, we’re now building v1.0 and inviting scientists from a wide range of fields to contribute scientific tasks.

🤔 What is SciVeri-Bench?
SciVeri-Bench is a deploy-first, scientist-driven benchmark for evaluating AI agents’ scientific verification capabilities on open-ended, novel scientific tasks.

🎯 Why are we building it?
- Stop to chasing maximing score in benchmark, rather we want to understand whether AI agents verify whether AI agent doing good science as domain scientist do
- Domain scientists only verify whether the science is good or bad, while agents do a lot of execution and exploration, but we really want to show that scientists can make novel discoveries

📝 How are tasks collected?
All task submissions and communication take place on OpenReview, so domain scientists can focus on proposing open-ended & novel scientific tasks and verifying agent outcomes.

🤝 What do scientists receive?
- All costs of running the agents are covered by the SciVeri-Bench team.
- Intellectual property rights to scientific discoveries made by AI agents belong to the scientist who proposed the task.
- Co-authorship on the SciVeri-Bench paper.

How can you contribute?
(1) Task Proposal: Propose a scientific task for SciVeri-Bench.
(2) Task Review: Review scientific tasks proposed by other scientists.

Interested in joining us?
👉 Sign up: forms.gle/bv9KZK6mW6ubBJ…
🌐 Website: sciveri-bench.com

🙏 Huge thanks to our team, contributing scientists, advisors, and sponsor!

AI researchers: @seungonekim @jinheonbaek @yoonholeee @JiayiiGeng

Task proposers: 16 scientists who have joined so far

AI advisors: @StevenDillmann @SoojungYang2 @AkariAsai @gneubig @wellecks @dongyeopkang

Sponsor: openrefinery.ai

We’re also looking for more sponsors and senior scientists!

Note: This work does not relate to any work conducted at Amazon