MaLiang-Harness: A Programmable Path to Image and Video Generation Paper • 2609.34309 • Published Sep 28 • 77
Towards Looped Models Done Right, Part II: Rethinking at Fixed Points Paper • 2610.06833 • Published Oct 5 • 26
Science Utopia? Closed-Loop LLM Simulation of Academic Research Ecosystems Paper • 2610.01257 • Published Oct 1 • 16
AutoSciBench: Autonomous Benchmark Generation for Evaluating Scientific Agents Paper • 2610.05140 • Published Oct 4 • 34
Memento 3: Model-Based Recursive Self-Improvement through Reflective Rulebooks Paper • 2610.11794 • Published Oct 8 • 38
In-Distribution Forcing for Long Video Generation at Test Time Paper • 2610.03120 • Published Oct 2 • 40
Memadapter: Counterfactual Adaptation Against Memory-induced Sycophancy Paper • 2610.05162 • Published Oct 4 • 49
SGF+: Decoupling Gradient Flows for Autoregressive Video Generation Paper • 2610.10429 • Published Oct 7 • 44
Science or Slop?: Benchmarking and Mitigating Scientific Slop in AI-Generated Papers Paper • 2610.00531 • Published Sep 30 • 30
VeriHarness: Scaling Agentic Verification for Long-Horizon Tasks Paper • 2610.00972 • Published Oct 1 • 21
PDE-JEPA: Predictive Representation Learning of Latent Dynamics Modeling for Parametric PDEs Paper • 2609.34715 • Published Sep 28 • 27
SearchJev: A Fast and Calibrated System-1 Model for Search Agents Paper • 2610.05107 • Published Oct 4 • 26
Questioning the Questions: Sustaining Self-Evolution in Reasoning Models Paper • 2610.04299 • Published Oct 3 • 49
GRACE: Generation-aware latent compression for efficient video generation Paper • 2610.10524 • Published Oct 7 • 65
Recursive Game Creator: An Agentic Product-Level Experience-Oriented Game Harness Paper • 2610.08621 • Published Oct 6 • 70
Scaling Trajectories for Complex Tasks through Recursive Self-Rewrite Paper • 2610.02826 • Published Oct 2 • 38