CHANNEL

Research

"Research" is a topic channel on AGI Hunt, an AI news site updated around the clock in real time. Coverage: Papers, new methods, empirical findings, technical and research reports, academia (incl. AI4Science); benchmark and dataset releases land here.

Daily roundup: the latest AI News Daily — the past 24 hours across the whole site, per channel and per company · browse the archive

• EASE: evidence-anchored spatial attention lifts multimodal RLVR by up to 3.1 points, EMNLP 2026 — jiqizhixin · 2026-09-11
• P=NP Explained: Why Class Schedules and Circuit Routing Are the Real Hard Problems — thesaraharminta · 2026-09-11
• Hypothesis: ASI Has a Mathematical Incentive to Preserve Human Diversity — No_Cause_2731 · 2026-09-11
• MutexaGPT: LLM agents plus MD simulations hit 40% on enzyme design, 4x the baseline — bravo_abad · 2026-09-11
• Single-author ECCV 2026 paper makes rolling shutter correction practical — ducha_aiki · 2026-09-11
• MetroLLM-Bench shows small fine-tuned models can match larger LLMs on transit-kiosk tasks — continker · 2026-09-11
• Report: OpenAI's internal model targets Riemann Hypothesis and P vs NP — 141_1337 · 2026-09-11(4 related)
• New essay argues scaling laws originate in data structure, not architectures — gleech · 2026-09-11
• CAROT: Optimal Transport-Based Token-Level Cross-Lingual Alignment Boosts Multilingual Accuracy by 11.2 Points — Bollegala · 2026-09-11
• Fruit flies trained to write: neural signals turn insect legs into a font generator — dejavucoder · 2026-09-11
• Stokes' theorem formalized in Lean 4 with true pullbacks and d²=0 proof — basedjensen · 2026-09-11
• Missing Benchmark for Agent Runtimes, Not Just Models, Reddit Thread Argues — Balance- · 2026-09-11
• PISA data preprocessed for AI agents: open dataset with AGENTS.md for instant analysis — pelayoarbues · 2026-09-11
• TempCloze Benchmark Tests Whether Video-LLMs Can Identify Missing Middle Video Segments — hkuhk · 2026-09-11
• Beihang's UniH3 Unifies Hierarchical Homogeneity and Heterogeneity for Medical Image Restoration — Beihang · 2026-09-11
• X-AuT: Progressive Audio-Encoder Compression for Speech LLMs via Cross-Scale Distillation — XPENG-AI · 2026-09-11
• RoMa v2 and LoMa to Be Presented at ECCV 2026 — ducha_aiki · 2026-09-11(2 related)
• Replication: LLMs can distinguish their own generated text from human text above chance — rohanpaul_ai · 2026-09-11
• Dev builds optical-flow diffusion model overnight to learn video motion — pixlpa · 2026-09-11(4 related)
• Two AACL-IJCNLP papers probe how fine-tuning reshapes LLM internals and why WiC stays hard — Bollegala · 2026-09-11
• Stanford Paper: Hybrid Local-Cloud Routing Cuts 60-80% of AI Costs — rohanpaul_ai · 2026-09-11(2 related)
• MLHC 2026 Proceedings Now Live on PMLR as Volume 340 — lawrennd · 2026-09-11
• LessWrong essay: a second-person stance toward Knightian uncertainty in AI alignment — LessWrong 精选 · 2026-09-11
• Dev Builds Local AI Memory System That Refuses to Hallucinate, Runs on One CPU Core — Unikum_01 · 2026-09-11
• Mathematician Daniel Litt releases video lectures on doing high-quality math with LLMs — caglar_ee · 2026-09-11
• More New Drugs Launched First in China in 2025 Than Anywhere Else, as AI Meets Speed — PatrickKidger · 2026-09-11
• PhyFilter: physics-informed filter lets sim-trained robots traverse real terrain without extra data — 机器之心 · 2026-09-11
• Giving a fly a voice: LLM turns fly brain intent into sentences — max_paperclips · 2026-09-11
• GPT-6 Pro produces candidate proof for Erdős problem #488, passing two arithmetic checkers — basedjensen · 2026-09-11
• UMO: unified in-context framework turns text2motion model into multi-task motion foundation — Michael_J_Black · 2026-09-11
• Clay Institute Changes Navier-Stokes Equation Status from Unsolved to Active — felpix_ · 2026-09-11
• First AI4AI Survey Maps Why AI Can't Yet Reliably Improve AI: The Composition Gap — 新智元 · 2026-09-11
• Independent kernel-level verification of OpenAI's Navier-Stokes Lean proof: all four builds pass — pvaa · 2026-09-11
• ACL Caps Authors to Save Reviewing, as Kyunghyun Cho Proposes ScholarCoin Token Economy — kchonyc · 2026-09-11
• OpenAI and Anthropic reportedly attacking P vs NP — what would a constructive P=NP proof break? — MohMayaTyagi · 2026-09-11
• Researcher warns CHI is about to be flooded by AI slop papers — IanArawjo · 2026-09-11
• Rumor: a counterexample already exists, Lean formalization being finalized — lpachter · 2026-09-11
• CiNet's 11th international conference on biological to general AI opens registration — kaixhin · 2026-09-11
• Looped Transformers Hit a Natural Ceiling: Most Test-Time Scaling Is Environment Interaction — generativist · 2026-09-11
• Classic Free Book From Python to NumPy Now Readable with AI Guidance — burkov · 2026-09-11(2 related)
• Author's caveat: stroke model's Dice 0.46 is an upper bound from outcome-fitted parameters — maier_ak · 2026-09-11
• Physics-Based Reaction-Diffusion Model Improves Stroke Lesion Prediction — maier_ak · 2026-09-11(2 related)
• Before Diffusion: Looking Back at VQGAN, the Pre-Diffusion Foundation of Image Generation — makeitrad1 · 2026-09-11
• Developer asks how to turn real codebases into instruction-to-code fine-tuning datasets — ImBadGuyInEveryStory · 2026-09-11
• UCSC adds DeepMind AlphaGenome Variant Impact scores for all 8.8B single-base changes — anshulkundaje · 2026-09-11
• Johns Hopkins releases EvoSafeHarness, evolving model- and domain-specific safety harnesses for agents — JohnsHopkins · 2026-09-11
• PhD Student Slams TMLR's Slow Turnaround: 2+ Months of Silence After Revision — Fantastic-Nerve-4056 · 2026-09-11
• Survey proposes PAPAV: a unified framework for multimodal embodied agents — MikeShou1 · 2026-09-11
• Terence Tao warns AI math capability strongly correlates with cyberattack ability — sytelus · 2026-09-11
• European Math Society hails OpenAI's Navier–Stokes solution, flags closed-model access concern — i_dg23 · 2026-09-11