Zhengyang Tang

Zhengyang Tang on Hugging Face Daily Papers: 18 papers, 5 in the top 3 of their day, 844 upvotes.

  1. JarvisHub: An Open Harness for Canvas-Native Multimodal Creative Agents 124 upvotes, #3 of 2026-07-28
  2. Training Open Models for Agentic Phone Use 16 upvotes, #17 of 2026-06-23
  3. GameCraft-Bench: Can Agents Build Playable Games End-to-End in a Real Game Engine? 55 upvotes, #5 of 2026-06-17
  4. PhoneHarness: Harnessing Phone-Use Agents through Mixed GUI, CLI, and Tool Actions 12 upvotes, #23 of 2026-06-16
  5. PhoneWorld: Scaling Phone-Use Agent Environments 8 upvotes, #44 of 2026-05-29
  6. Safe, or Simply Incapable? Rethinking Safety Evaluation for Phone-Use Agents 2 upvotes, #63 of 2026-05-12
  7. Claw-Eval-Live: A Live Agent Benchmark for Evolving Real-World Workflows 40 upvotes, #6 of 2026-05-01
  8. Cut Your Losses! Learning to Prune Paths Early for Efficient Parallel Reasoning 23 upvotes, #7 of 2026-04-20
  9. Do Phone-Use Agents Respect Your Privacy? 9 upvotes, #17 of 2026-04-02
  10. Kimi K2.5: Visual Agentic Intelligence 219 upvotes, #2 of 2026-02-03
  11. CALM Before the STORM: Unlocking Native Reasoning for Optimization Modeling 19 upvotes, #16 of 2025-10-09
  12. CoRT: Code-integrated Reasoning within Thinking 18 upvotes, #9 of 2025-06-12
  13. Learning from Peers in Reasoning Models 41 upvotes, #4 of 2025-05-13
  14. RealCritic: Towards Effectiveness-Driven Evaluation of Language Model Critiques 29 upvotes, #3 of 2025-01-27
  15. Enabling Scalable Oversight via Self-Evolving Critic 66 upvotes, #1 of 2025-01-13
  16. Omni-MATH: A Universal Olympiad Level Mathematic Benchmark For Large Language Models 26 upvotes, #6 of 2024-10-15
  17. MathScale: Scaling Instruction Tuning for Mathematical Reasoning 13 upvotes, #6 of 2024-03-06
  18. Synthetic Data (Almost) from Scratch: Generalized Instruction Tuning for Language Models 52 upvotes, #2 of 2024-02-21

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.