2026
10,780 articles · 136 days · 10 categories · avg 79 per day
Tsinghua's LoopWAM Tops RoboChallenge Table30 and Goes Open Source
LoopWAM, a world action model from Tsinghua, 原力灵机 and 极寰时代, took first place on the RoboChallenge Table30 benchmark. The code, paper and project page are already…
Lean Compilation Overstates Formalization Accuracy by Up to 29.5 Points
An agent with Lean tools compiles 87.2% of statements, but only 57.7% are judged correct. The benchmark authors measured how much compilation overestimates…
TTR-Compressor Compresses Sequences with Regression and Cuts Visual Tokens by 55.6%
The new TTR-Compressor treats token compression as a regression problem and uses its coefficients as the compressed representation: same quality with 55.6% fewer…
EinsiaAI Launches Quant Arena: Half of 9 LLMs Lose Money on China's A-Share Market
ReGDiff Teaches Diffusion Models to Design Metamaterials Without Copying or Producing Defects
Mechanistic Uncertainty in LLMs: Measuring Causal Robustness of Internal Mechanisms
8B-Parameter Model Beats Claude Opus 4.8 at Generating Scientific Ideas
LEASH hands hard steps to a strong agent and cuts Opus 5 costs by 19.2%
LEASH puts a cheap agent in the driver's seat and calls Claude Opus 5 only at hard moments: 73.66% on OSWorld at a cost 19.2% lower.
Volcano Engine Opens Seedance 2.5 4K to 20 Studios for Film Projects
Volcano Engine has launched an enterprise pilot of Seedance 2.5 4K. The first access batch went to 20 studios and directors, with big-screen cinema as the stated…
Why You Can't Trust an AI Agent to Grade Its Own Work: A Report Claiming -9,500 Lines When the Codebase Actually Grew by +21,489
Lean4AgenticRL: Multi-Level Process Rewards Push LAR-27B to 77.3%
HeirCommit Filters Out Up to 83% of Harmful Updates in Self-Evolving LLMs
ACT3D teaches a VLM to explore 3D scenes and beats Gemini 3.6 Flash: 53.2% across six benchmarks
Lovable launches Deploy on Azure: your apps and data live in your own subscription
Lovable opened a private preview of Deploy on Azure: apps, database and logs run in the customer's subscription on AKS and PostgreSQL, while the editor and agent…
Google Antigravity Finds an Antimicrobial Peptide Candidate in a 97-Amino-Acid Protein
Google's Antigravity agent platform and the ProtNLM model found a new 97-amino-acid bacteriocin in UniProt's "dark matter." It is now being tested in the lab.
Mistral releases Timestral: a 209M-parameter model that tops three benchmarks
Mistral's first time-series model family ranks first on GIFT-Eval, TIME and fev-bench with 209M parameters, and it runs on an ordinary server CPU.
RVCBench Tests 18 Voice Cloning Models on Noise, Compression, and Anti-Cloning Protection
The Expectations Gap: 91% of Anthropic Employees and 39% of Americans Expect a Nobel-Level AI Breakthrough
Anthropic: AI Agents Now Outnumber Employees 6 to 1, and Junior Hiring Is Shrinking
LILIE teaches image generation through language, cutting gFD from 66.84 to 53.01
Competence Over Accuracy: Why GPT and Qwen Lose Logic When Problems Get Rephrased
A new methodology called Competence-Based Evaluation (CBE) has uncovered a fundamental flaw in LLM logical reasoning: even strong GPT and Qwen models fail on the…
A no-hair theorem for transformers: the prompt is forgotten as t^-α, α≈0.55–0.7
After t tokens, the distance between hidden states of different prompts decays as t^-α. Trained models show α≈0.55–0.7; the proven bound is O(P(log t)^{L-1}/t).
MOSS-Image: Open 5B Model Renders Long Text with 96.3% Coverage
MIST shows video world models don't forget; they fail to carry out the command
"Wait" in Reasoning Models Turns Out to Be Inherited from Data, Not Discovered by RL
Vercel Shop Reaches GA: Shopify Storefronts Learn to Sell to AI Agents
HumanoidPose Estimates Humanoid Robot Pose From Video, Cutting Joint-Angle Error by 18–32%
HumanoidPose recovers humanoid robot joint angles from video alone. Trained on synthetic data for 16 robot models, it cut error by 18–32% against the best baseline…
Researchers Annotate 2,122 WildChat Conversations: 71.1% of Legal Chats Seek Personal Advice
A taxonomy of legal queries to LLMs, 2122 WildChat-1M conversations and open models: 71.1% of chats with legal content ask for personalized or practical advice.
StopLoss Cuts Off Doomed Coding-Agent Runs and Saves Up to 48% of Costs
StopLoss, a small external monitor, predicts agent failure from the latest steps and halts the run: 4.5% lower cost on Verified and up to 31% on SWE-bench Pro.
Changing the definition of "unsafe" moves a guardrail score from 0 to 100 with 0.03% of weights
Anthropic to put 5% of its shares into a Public Benefit Fund, with the stake growing as the company grows
Researchers Propose Anti-Persistence: LLM Agent Memory Without Stored Rules
