Introducing RSI-Anything: Unlocking Recursive Self-Improvement with World Models 🚀 Can AI learn to improve itself by predicting the consequences of its actions? Built on JEPA-Anything, RSI-Anything uses world models to guide action selection and learning. Feedback improves the policy or workflow and, when needed, the world model. Validated updates carry forward to shape the next round of experience. Across code, GUI interaction, laboratory planning, and reaction optimization: • Code-repair success: 26.3% → 31.2%, with the coder fixed • GUI agent task: 39.4% fewer actions, with higher completion • Lab-planning scores: 21.80% → 46.01% over five policy versions • Mean best observed reaction yield: 96.15% vs. 82.16% for Bayesian optimization, under the same 40-experiment budget 📄 Paper: arxiv.org/abs/2610.13912 💻 Code: github.com/Gen-Verse/RSI-Any… Watch the video below 👇 #RSI #WorldModels #RSI #AIAgents #AIforScience #JEPA

Excited to introduce DiscoveryRSI! One model that recursively learns to discover valuable hypotheses, and gets sharper with every research round. Let's RSI Discovery! 📄 Paper: arxiv.org/abs/2610.23839 💻 Code: github.com/Gen-Verse/Discove… #RSI #AIforScience #AIAgents #ScientificDiscovery

Recursive self-improvement is one of the most discussed ideas in AI right now. I am particularly interested in how it applies to science. I saw this really cool new paper with neat ideas, especially on hypothesis generation. A good read. Specifically, they propose DiscoveryRSI, which applies RSI to scientific hypothesis generation in four life-science domains. Its 27B model produced hypotheses that match research published after its training cutoff 6.3% of the time, compared with 2.5% for GPT-6 and 2.3% for Claude Opus 5. The system grows trees of hypotheses with multi-agent evolutionary search and turns the chosen investigations into executable tasks. After each run, it assigns credit to the hypotheses that contributed, replays the research harness to find productive strategies and repeated failures, and applies accepted improvements to new investigations. Automated hypothesis quality rose from 4.86 to 5.39 after two rounds, with the largest gains in falsifiability. arxiv.org/abs/2610.23839 Chat with Paper: academy.dair.ai/papers/disco…

Can research agents get better at science by learning from the research they do? Introducing DiscoveryRSI 🧬: a framework that turns hypothesis-driven research into a recursive self-improvement loop. 📄 Paper: arxiv.org/abs/2610.23839 💻 Code: github.com/Gen-Verse/Discove… 🔴 Live Arena: foundry-arena.rsi-arena.work… 🔹 What is a valuable hypothesis? We define it as a joint object H = (claim, investigation plan, decision criteria, alternative explanations). A hypothesis is valuable if resolving it would change a research decision. It is testable if the plan can produce evidence that supports, weakens, or bounds the claim. 🔹 How it works - Lead, Generation, Evolution, Critique and Progress agents grow hypothesis trees through evolve / repair / crossover / restart - Every version stays linked to its evidence, data, assessments and contribution records - Selected plans compile into executable Harbor tasks 🔹 Three self-improvement mechanisms ① Hypothesis credit optimization: blinded TrueSkill assessments assign bounded credit to the agent steps that actually contributed, and DiscoveryRSI-27B is trained only on those steps (Qwen3.8-27B init, 256K context; 2,397 of 15,792 trajectories kept; 2.12% of tokens supervised) ② Harness replay: diagnose recurring failures such as repeated exploration, failed tool use, lost artifacts and premature closure, then refine the workflow ③ Rediscovery: run the accepted model and harness updates on fresh questions 🔹 Results (DFM-Bench, 4 life-science domains) - Hypothesis quality 4.86 → 5.18 → 5.39 over two rounds - Falsifiability 4.00 → 6.20, alternative explanations 4.44 → 5.98, data resolvability 3.73 → 4.99 - Evolve/repair share 77.5% → 91.7%, restarts 21.7% → 5.0%: the agent learns to keep developing an idea instead of starting over - Caveat: importance and novelty dip slightly. The model learns to design better tests, not yet to choose bolder questions. 🔹 Expert review 27 physicians and bioscience researchers scored 292 anonymized proposals: - 41% of DiscoveryRSI-27B revisions improve quality vs 31% (GPT-5.6 Sol) and 25% (GPT-6 Luna) - Only 12% make things worse vs 32% and 31% 🔹 Does it anticipate real science? Against 16,440 publications filtered by each model's cutoff, 6.3% of DiscoveryRSI-27B's hypotheses connect to later published research (16.9% in therapeutic targets). Comparison: DeepSeek 5.4%, GLM 2.8%, GPT 2.5%, Claude 2.3%. Watch six models do research in real time in DiscoveryRSI Arena, and follow every proposal, critique and compile step 👇 foundry-arena.rsi-arena.work… Co-led by Jianyuan Zhong, Chenwei Wu & Zhaochen Yu, with wonderful collaborators across PhAI Labs, CUHK, Stanford, MIT, Harvard Medical School, Oxford, Princeton, PKU and more. Grateful to @james_y_zou @Jiaru_Zou @Charles_Y_Wu @zhenfei_yin_yzf 🙏

🚀 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/Discove… 🔴 foundry-arena.rsi-arena.work…

SGLang became the infra for serving LLMs. @PhAILabs is building the infra for AI-driven science. Proud to see PhAI support the @sgl_project community at SGLang Summit — open infrastructure is how both AI and science scale. 📅 Nov 12–13 · Fort Mason, SF

We're incredibly proud to support and join the @sgl_project community at the SGLang Summit. @PhAILabs Can't wait to connect with everyone there!

Excited to introduce RSI-Anything! One RSI model that works across coding, GUI, planning, or even Science tasks. Let's RSI Everything! 📄 Paper: arxiv.org/abs/2610.13912 💻 Code: github.com/Gen-Verse/RSI-Any… #RSI #WorldModels #RSI #AIAgents #AIforScience #JEPA

This Friday (Oct 9) at #COLM2026, we're hosting the 2nd Workshop on Lifelong Agents: a full day on how to build agents that keep learning after deployment while staying reliable and aligned. Speakers: • Sergey Levine (UC Berkeley) • Graham Neubig (CMU / OpenHands) • Yu Su (NeoCognition / Ohio State) • Fred Sala (Snorkel AI / UW–Madison) • Philippe Laban (Microsoft Research) • Shi Dong (RadixArk) • Ying Liu (Scale AI) 📍 Continental Ballroom 6, Hilton San Francisco Union Square 🗓 Fri, Oct 9 · 8:30 AM – 5:00 PM PT 💻 Virtual option available RSVP: luma.com/h6vm3nqc Schedule: lifelongagent.github.io/ Thanks to my co-organizers, our reviewers and authors, and to Salesforce AI Research, Snorkel AI, and Tencent for helping cover registration for student authors! If you're at COLM and working on agents, come say hi! #AIAgents #ContinualLearning

JEPA is bigger than vision. Our JEPA-Anything explores predictive representation learning across seven domains, from biology to weather. It takes a step toward general-purpose world models. Paper: arxiv.org/abs/2609.20800 Code: github.com/Gen-Verse/JEPA-An…

Excited to introduce JEPA-Anything! One world model that works across molecules, cells, fluids, patients, robots, or even weather. Let's World Model Everything! 📄Paper: arxiv.org/abs/2609.20800 💻Code: github.com/Gen-Verse/JEPA-An…

Excited to share that GenEnv and Skill-Entropy have both been accepted to NeurIPS 2026! 🎉 Two connected directions toward continually improving agents: 🔹 GenEnv co-evolves agents and environment simulators, generating tasks whose difficulty adapts to the agent’s current capabilities. 📄 arxiv.org/abs/2512.19682 💻 github.com/Gen-Verse/GenEnv 🔹 Skill-Entropy measures the difficulty of switching between reasoning skills and turns it into an RL signal for improving long-horizon reasoning. 📄 arxiv.org/abs/2608.05139 💻 github.com/Gen-Verse/Skill-E… We’ve also brought techniques from both into ScienceBuddy, extending this research toward scientific agents that improve through researcher interaction, harness evolution, and model RL. 🔬 ScienceBuddy: Recursive-in-Recursive Self-Improvement for Interactive Scientific Agents 📄 arxiv.org/abs/2609.17523 💻 github.com/Gen-Verse/Science… Our broader goal: use challenging scientific workflows to push the frontier of agent capabilities and discovery intelligence. Huge thanks to all our collaborators! Try ScienceBuddy: science-buddy.io #NeurIPS2026 #RSI #AIforScience

Must-read papers of the week ▪️ JEPA-Anything ▪️ Modality-Autoregressive World-Action Models ▪️ In-Context Robot Learning with VLM Agents ▪️ Dream-RSI: Recursive Self-Improvement through Evolving Worlds ▪️ ModularRSI: Modular and Generalizable Recursive Harness Self-Improvement ▪️ DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression ▪️ SoL-Pi: Recursively Scaling Auto-Research Loops for Efficient Agent Harness ▪️ Confidence Comes from Experience: Experiential Confidence Estimation from Reasoning to Agents ▪️ AliceAI-Foundation-80B-A3B-Base (model release) ▪️ Video DeltaNet: A Video-Native Hybrid Attention for Livestream Video Generation ▪️ ScienceBuddy: Recursive-in-Recursive Self-Improvement for Interactive Scientific Agents ▪️ World Modeling in Transformers ▪️ When2Think: Learning Difficulty-Aware Length Control for Efficient Hybrid Reasoning Models Explore these to keep up with main AI trends. Here’s also the full list of stunning papers + links and our weekly AI news digest: turingpost.com/p/tracktwo-ai

Skill Entropy is accepted to #NeurIPS2026! 🎉🥳

Can scientific agents get better by working alongside researchers? With ScienceBuddy, we’re exploring how feedback from real research can drive both harness refinement and model training—and push the frontier of discovery intelligence. Try it free: science-buddy.io

Introducing ScienceBuddy — a free workspace for scientific agents that improve through researcher collaboration. Use GPT-6 in ScienceBuddy at no cost. GPU-accelerated, and fused with the JEV framework. 🧵 Two loops: 🔹 Inner loop — refines the agent harness 🔹 Outer loop — trains the model with rubric-guided RL Together: Recursive-in-Recursive Self-Improvement. ScienceBuddy explores how scientific agents can improve through sustained collaboration with researchers. 🔬Try it free: science-buddy.io #ScienceBuddy #PhAILabs #AI4Science

Today, we're launching ScienceBuddy — a free workspace for scientific agents. Get GPT-6 at no cost, fully GPU-accelerated, and fused with our JEV framework. The engine behind it relies on two loops: 🔹 Inner loop — refines the agent harness as you collaborate 🔹 Outer loop — trains the model via rubric-guided RL Together, it drives Recursive-in-Recursive Self-Improvement. We genuinely want to see how agents improve through sustained, real-world collaboration. Try it out, push its limits, and let's build better AI for science together! 🔬 science-buddy.io #ScienceBuddy #PhAILabs #AI4Science

Meet Jev for Science. Same biochemistry question. Wildly different speeds. ❌ ScienceBuddy: 7.00s (still processing) ✅ ScienceBuddy-Jev: 0.62s (Done) Read a 58k-character paper. Nailed the correct answer. 99.971% confidence. Experience the next level of scientific retrieval and reasoning. Test it live: science-buddy.io #AI #Research

Introducing Jev for Science. We ran a tough plant biochemistry question: ⏱️ ScienceBuddy-Jev: 0.62s. Read a 58k-char paper, selected B (Correct) at 99.971% confidence. ⏱️ ScienceBuddy: 7.00s, still on model call 2/6. It retrieves, judges, and answers with perfectly calibrated confidence. Try it now: science-buddy.io

"JEPA-Anything: Learning Predictive Models across Different Worlds" This research proposes one world-modeling framework that works across vision, biology, control, molecules, physics, weather, and clinical data. They used something called Orthogonal Predictive Factorization, which splits a single JEPA latent state into complementary factors that predict different parts of the world and recombine into a complete state. This improves all 10 matched dynamics tasks and even produces factors that recover Kepler’s scaling law and nominate a biologically validated intervention. alphaxiv.org/abs/2609.20800