Daily Papers of 2026-04-20

  1. Elucidating the SNR-t Bias of Diffusion Probabilistic Models 73 upvotes, #1 of 2026-04-20
  2. DiPO: Disentangled Perplexity Policy Optimization for Fine-grained Exploration-Exploitation Trade-Off 61 upvotes, #2 of 2026-04-20
  3. Maximal Brain Damage Without Data or Optimization: Disrupting Neural Networks via Sign-Bit Flips 57 upvotes, #3 of 2026-04-20
  4. Qwen3.5-Omni Technical Report 56 upvotes, #4 of 2026-04-20
  5. PersonaVLM: Long-Term Personalized Multimodal LLMs 46 upvotes, #5 of 2026-04-20
  6. Web Retrieval-Aware Chunking (W-RAC) for Efficient and Cost-Effective Retrieval-Augmented Generation Systems 26 upvotes, #6 of 2026-04-20
  7. Mind DeepResearch Technical Report 23 upvotes, #7 of 2026-04-20
  8. Cut Your Losses! Learning to Prune Paths Early for Efficient Parallel Reasoning 23 upvotes, #7 of 2026-04-20
  9. Where does output diversity collapse in post-training? 22 upvotes, #9 of 2026-04-20
  10. Motif-Video 2B: Technical Report 20 upvotes, #10 of 2026-04-20
  11. (1D) Ordered Tokens Enable Efficient Test-Time Search 18 upvotes, #11 of 2026-04-20
  12. QuantCode-Bench: A Benchmark for Evaluating the Ability of Large Language Models to Generate Executable Algorithmic Trading Strategies 15 upvotes, #12 of 2026-04-20
  13. TIPSv2: Advancing Vision-Language Pretraining with Enhanced Patch-Text Alignment 12 upvotes, #13 of 2026-04-20
  14. Repurposing 3D Generative Model for Autoregressive Layout Generation 12 upvotes, #13 of 2026-04-20
  15. Learning Adaptive Reasoning Paths for Efficient Visual Reasoning 8 upvotes, #15 of 2026-04-20
  16. Can Large Language Models Reinvent Foundational Algorithms? 7 upvotes, #16 of 2026-04-20
  17. RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies 7 upvotes, #16 of 2026-04-20
  18. The Amazing Agent Race: Strong Tool Users, Weak Navigators 7 upvotes, #16 of 2026-04-20
  19. AccelOpt: A Self-Improving LLM Agentic System for AI Accelerator Kernel Optimization 4 upvotes, #19 of 2026-04-20
  20. Universal statistical signatures of evolution in artificial intelligence architectures 4 upvotes, #19 of 2026-04-20
  21. PRL-Bench: A Comprehensive Benchmark Evaluating LLMs' Capabilities in Frontier Physics Research 4 upvotes, #19 of 2026-04-20
  22. EdgeDetect: Importance-Aware Gradient Compression with Homomorphic Aggregation for Federated Intrusion Detection 3 upvotes, #22 of 2026-04-20
  23. NTIRE 2026 Challenge on Video Saliency Prediction: Methods and Results 3 upvotes, #22 of 2026-04-20
  24. GTA-2: Benchmarking General Tool Agents from Atomic Tool-Use to Open-Ended Workflows 3 upvotes, #22 of 2026-04-20
  25. ArtifactNet: Detecting AI-Generated Music via Forensic Residual Physics 3 upvotes, #22 of 2026-04-20
  26. VEFX-Bench: A Holistic Benchmark for Generic Video Editing and Visual Effects 3 upvotes, #22 of 2026-04-20
  27. Hierarchical Codec Diffusion for Video-to-Speech Generation 2 upvotes, #27 of 2026-04-20
  28. TwinTrack: Post-hoc Multi-Rater Calibration for Medical Image Segmentation 1 upvotes, #28 of 2026-04-20

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.