Daily Papers of 2026-08-24

  1. Graph Engineering in the Era of LLM Agents: From Individual Intelligence to System Intelligence 61 upvotes, #1 of 2026-08-24
  2. Let's Scale Step by Step: Compute-Efficient Hyperparameter Transfer for Large-Scale Mixture-of-Experts 46 upvotes, #2 of 2026-08-24
  3. ParaTempo: Efficient Parallel Reasoning via Temporal Confidence 38 upvotes, #3 of 2026-08-24
  4. InfinityEdit: Infinite Video Editing with a Lightweight Edit-Ignition Adapter 38 upvotes, #3 of 2026-08-24
  5. Beyond Correctness: Benchmarking and Aligning Response Behaviors in Hybrid-Thinking MLLMs 35 upvotes, #5 of 2026-08-24
  6. OmniAssistBench: Assistant-style Interaction Benchmark for Omni-LLMs 30 upvotes, #6 of 2026-08-24
  7. UniSpace: Unified Visual Representation and Scalable Multimodal Modeling 15 upvotes, #7 of 2026-08-24
  8. Every Coin Has Two Sides: On the Dual Nature of Generalization in On-Policy Distillation of Large Language Models 15 upvotes, #7 of 2026-08-24
  9. WorldMind: Decoupled Game World Model for State-Aware NPC Behavior 15 upvotes, #7 of 2026-08-24
  10. AgentMercury: Your Agent Can Synthesize Verifiable Environments for Business Scenarios at scale 12 upvotes, #10 of 2026-08-24
  11. EviRank: Structured Relevance Evidence for Multimodal Image Re-ranking 11 upvotes, #11 of 2026-08-24
  12. Hydra-0: Action Flow for Generalist World Modeling and Control 10 upvotes, #12 of 2026-08-24
  13. Human-Centric Intelligence in the Era of Foundation Models: A Survey 9 upvotes, #13 of 2026-08-24
  14. Partition the Support, Reconstruct the Residual: Training-Free Sparse Attention for Video Generation and World Models 9 upvotes, #13 of 2026-08-24
  15. Daedalus-150M: A Convolution-Attention Hybrid Designed for CPU Inference 8 upvotes, #15 of 2026-08-24
  16. Llama-Mobile: Efficient 2.7-Bit Quantization of VLMs 8 upvotes, #15 of 2026-08-24
  17. Towards Faithful Simulation of Human Shopping Behavior 6 upvotes, #17 of 2026-08-24
  18. CLEAR: Continuous Latent Adapter Routing for Utility-Preserving LLM Safety Alignment 5 upvotes, #18 of 2026-08-24
  19. Peer-Voted LLM-Agent Stress Tests Find Feed-Induced Lexical Convergence but No Reliable Matched-Exposure Advantage for Distributed Sources 4 upvotes, #19 of 2026-08-24
  20. PhysCaP: Grounding Code-as-Policy Agent with Physics-Informed Exploration 4 upvotes, #19 of 2026-08-24
  21. FlavourBench: Ranking Frontier Language Models with Executable Culinary Ground Truth 3 upvotes, #21 of 2026-08-24
  22. Hadith computational science in the age of large language models: a critical narrative review 2 upvotes, #22 of 2026-08-24

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.