Daily Papers of 2026-08-14

  1. Alaya-EVOKE: From Linear-Scaling Supervision to Endless World 132 upvotes, #1 of 2026-08-14
  2. DarwinX: Evolving Agent Harnesses Through Natural Selection 110 upvotes, #2 of 2026-08-14
  3. LLMRouter: Unified Infrastructure for Developing, Evaluating, and Deploying LLM Routers 108 upvotes, #3 of 2026-08-14
  4. DreamX-Phi 1.0: Action-Conditioned Video World Model for Robotic Manipulation 98 upvotes, #4 of 2026-08-14
  5. OmniScientist: An Omni-Modal Omni-Discipline AI Scientist 87 upvotes, #5 of 2026-08-14
  6. Intern-S2-Preview: Scientific Agentic Foundation Model 68 upvotes, #6 of 2026-08-14
  7. AutoDesign: Meta-Harness Optimization for Long-Horizon Agentic Design 53 upvotes, #7 of 2026-08-14
  8. How Can Rhetoric Reward-Hack AI Reviewers? Dissecting Rhetorical Sensitivity in AI-Based Peer Review 48 upvotes, #8 of 2026-08-14
  9. PlayWorld: Benchmarking World Models with Agent Players over Long-Horizon Objectives 45 upvotes, #9 of 2026-08-14
  10. Spatial Memory Agent: Experience-Grounded Procedure Memory for Spatial Intelligence 43 upvotes, #10 of 2026-08-14
  11. Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus 30 upvotes, #11 of 2026-08-14
  12. LiveAnimate: Stable Long-Form Streaming Human Animation in Real-Time 23 upvotes, #12 of 2026-08-14
  13. Full-bandwidth transformer 22 upvotes, #13 of 2026-08-14
  14. UniSwap: Streaming Audio-Visual Identity Swapping for Talking Videos 22 upvotes, #13 of 2026-08-14
  15. Context-Matched Distillation: Teacher Causality for Autoregressive Video Distillation 18 upvotes, #15 of 2026-08-14
  16. H2R-Bench: Benchmarking Human-to-Robot Manipulation Video Generation in World Models 17 upvotes, #16 of 2026-08-14
  17. An AI4AI Framework for Visual Token Pruning 15 upvotes, #17 of 2026-08-14
  18. Thought-Level Beam Search for Reasoning 15 upvotes, #17 of 2026-08-14
  19. SKILLER: Language-Level Reinforcement Learning for Reusable Skill Extraction in Small Language Models 15 upvotes, #17 of 2026-08-14
  20. Knowing When to Quit: Diagnosing and Training LLMs to Abort Futile Reasoning 14 upvotes, #20 of 2026-08-14
  21. Maglev: Sliding Recurrent Memory 14 upvotes, #20 of 2026-08-14
  22. LycheeMemory V2: Efficient Long-Term Memory for LLM Agents via Semantic Segment-Level Consolidation 14 upvotes, #20 of 2026-08-14
  23. Are You Sure You're Sure? On the Impact of Instruction Tuning on Confidence and Lexical Diversity 12 upvotes, #23 of 2026-08-14
  24. Hybrid-Policy Self-Editing for Composable Unstructured Knowledge Editing 11 upvotes, #24 of 2026-08-14
  25. From Inaudible Inputs to Model Failures: Low-Frequency Safety Risks in LALMs 10 upvotes, #25 of 2026-08-14
  26. Specification-first convergence with an AI coding agent: a case study of dismantling a core architectural invariant across 189 files in a 717k-line codebase with no test oracle and no human code review 10 upvotes, #25 of 2026-08-14
  27. PixSDS: Why Latent SDS Makes Noisy Pixels 10 upvotes, #25 of 2026-08-14
  28. RibAssist 3D: Biplanar Rib-Fracture Detection, Addressing, and Selective 3D Localization from CT-Derived Projections 9 upvotes, #28 of 2026-08-14
  29. Mitigating Gender Bias in English to Romanian Machine Translation 9 upvotes, #28 of 2026-08-14
  30. TailBooster: A Dual-Layer Generative Framework for Extreme Value Augmentation with Operational Validity Enforcement 8 upvotes, #30 of 2026-08-14
  31. CW-BASS v2: Saturation-Aware Pseudo-Label Selection for Semi-Supervised Segmentation under Foundation-Model Teachers 8 upvotes, #30 of 2026-08-14

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.