Daily AI Papers — August 12, 2026

1. On-Policy Self-Distillation without Any Supervision

Authors: Yijiang Li, Bingyang Wang, Yijun Liang, Yunjie Tian, Di Fu, Nuno Vasconcelos arXiv: arxiv.org/abs/2608.06296 Summary: On-policy (Self-)Distillation (OPD / OPSD) has shown strong potential for post-training large language models (LLMs). However, existing methods still rely heavily on external supervision, including ground-truth signals, environmental feedback, or guidance from larger models, and therefore fall short of genuine “self”-distillation. Trending because: 183 HuggingFace upvotes + one of the most-upvoted papers in today’s feed.

2. ComBodied Agents: a New Paradigm of Human-Centric Agentic AI

Authors: Qianggang Ding, Xingyao Wang, Rui Feng, Zhibin Wang, Feixiang Wang, Kelong Mao, Hao Sun, Zhiyao Luo, Jiankai Tang, Lei Li, Jiadong Guo, Minheng Ni, Weicong Lin, Chenxi Yang, Hongxiang Gao, Zhenghua Chen, Yang Bai, Min Wu, Jun Cheng, Huazhu Fu, Dacheng Tao, Bang Liu arXiv: arxiv.org/abs/2608.10915 Summary: After an older adult misses a medication dose, a software agent can send another reminder and an embodied agent can bring the medication. Yet neither explains whether the person forgot, is confused, has side effects, or deliberately refused, nor what support is appropriate. Trending because: 148 HuggingFace upvotes + one of the most-upvoted papers in today’s feed.

3. Co-Evolution in Agentic Systems: Toward Self-Directed Evolution Beyond Human Design

Authors: Qing Zong, Jiayu Liu, Junhao Shen, Zecong Tang, Linsi Wu, Yuxuan Liu, Rui Wang, Zhaowei Wang, Weiqi Wang, Cheng Qian, Xiusi Chen, Yangqiu Song arXiv: arxiv.org/abs/2608.10299 Summary: Agentic systems are increasingly expected to improve after deployment, yet single-entity self-evolution is often bounded by a static learning context, such as fixed tasks and feedback. This survey focuses on co-evolution in agentic systems, a multi-component form of self-evolution in which multiple agents and their environment impose adaptive pressure on one another. Trending because: 106 HuggingFace upvotes + one of the most-upvoted papers in today’s feed.

4. Beyond Pixels: From Video Priors to 4D Worlds

Authors: Zihao Liu, Xiaolong Shen, Zhenglin Zhou, Ruijie Quan, Yi Yang arXiv: arxiv.org/abs/2608.10744 Summary: 4D generation synthesizes dynamic 3D scenes from conditions such as text or images. Existing methods either reconstruct generated RGB videos with a separate 4D model or adapt a particular video generator to predict geometry directly. Trending because: 100 HuggingFace upvotes + one of the most-upvoted papers in today’s feed.

5. Articulated Object Reconstruction from Rest-State Observation

Authors: Daeun Lee, Jaeah Lee, Woosung Kim, Haebeom Jung, Jaesik Park arXiv: arxiv.org/abs/2607.27749 Summary: Building interactive digital twins requires recovering both 3D geometry and the kinematic structures that govern how objects articulate. Yet existing methods for articulated object reconstruction require explicitly observable motion from multiple articulation states. Trending because: 33 HuggingFace upvotes + one of the most-upvoted papers in today’s feed.

6. AdvFD: Boosting Visual Generation via Adversarial Fr’echet Distance Loss

Authors: Mingju Gao, Jingkai Zhou, Kun Gai, Changqian Yu, Hao Tang arXiv: arxiv.org/abs/2608.11205 Summary: Fréchet distance has recently emerged as an effective distribution-level objective for generator post-training, complementing the conventional sample-level diffusion and flow-matching losses. However, directly optimizing Fréchet objectives can cause Fréchet hacking. Trending because: 16 HuggingFace upvotes + trending in today’s HuggingFace papers feed.

7. Mendel Gödel Machine: Recursive Self-Improving Coding Agents via Comparative Evolution

Authors: Changzhi Liu, Yilun Liu, Sikuan Yan, Volker Tresp, Yunpu Ma arXiv: arxiv.org/abs/2608.07645 Summary: Self-improving coding agents that iteratively rewrite their own source code have demonstrated impressive performance on coding tasks. However, existing solutions generally derive self-modification from a single failure trajectory at a time, overlooking rich comparative signals available in the agent’s expanding archive of past attempts. Trending because: 15 HuggingFace upvotes + trending in today’s HuggingFace papers feed.

8. Business Arena: Benchmarking LLM Agents in a Realistic Marketplace

Authors: Yijun Pan, Yukun Lian, Kunyu Shi, Junbo Li, Hongwei Xue, Sicong Xie, Guannan Zhang, Xiaoying Xing arXiv: arxiv.org/abs/2608.08621 Summary: Running a business is a challenging form of intelligent work. Operators must infer opportunities from partial signals, commit capital under uncertainty, adapt to delayed outcomes in a changing market, and satisfy regulatory obligations before trading legally. Trending because: 13 HuggingFace upvotes + trending in today’s HuggingFace papers feed.

9. RynnValue: Scaling Robotic Value Foundation Models with Temporal Distance

Authors: Dongchi Huang, Hongyin Zhang, Bohan Hou, Siteng Huang, Zhian Su, Hang Guo, Tong Lu, Zhaofeng Xu, Jiahao Tang, Jianfei Yang, Donglin Wang, Peixi Peng, Mingxiu Chen, Deli Zhao, Xin Li arXiv: arxiv.org/abs/2608.09853 Summary: General-purpose reward models are increasingly the bottleneck for scaling robot learning, yet the recipe for learning value-related capabilities from large-scale heterogeneous corpora remains underexplored. Existing approaches tie supervision to task-internal anchors such as preferences or normalized progress, none of which transfer cleanly across embodiments and data sources. Trending because: 11 HuggingFace upvotes + trending in today’s HuggingFace papers feed.

10. VibeLifeBench: Can Your Life Agent Be Proactive and Persistent in a Living World?

Authors: Xiaohongshu Inc arXiv: arxiv.org/abs/2608.10875 Summary: Large language model (LLM) agents are increasingly deployed as personal assistants. Existing evaluations, however, mostly use short, self-contained requests in static environments. Trending because: 11 HuggingFace upvotes + trending in today’s HuggingFace papers feed.

11. Ex-Omni-2D: Expressive Omni-Modal Dialogue Models with Native Visual Presence

Authors: Haoyu Zhang, Zhipeng Li, Xiaoying Tang, Tianshu Yu, Yiwen Guo arXiv: arxiv.org/abs/2608.10720 Summary: Omni-modal dialogue models can understand multimodal inputs and synthesize spoken replies, yet their responses remain visually disembodied. We introduce Ex-Omni-2D, an omni-modal dialogue framework that generates a coordinated response comprising text, personalized speech, and reference-conditioned video. Trending because: 9 HuggingFace upvotes + trending in today’s HuggingFace papers feed.

12. A Hybrid Nested Harness for Decoupling Structure and Parameters in LLM-Driven Optimization

Authors: Víctor Gallego arXiv: arxiv.org/abs/2608.08156 Summary: In evolutionary algorithms powered by language models, the LLM acts as a single operator that simultaneously updates structural components (like control flow) and continuous parameters. While LLMs can be good at the first, they are not efficient at the second, wasting tokens taking discrete jumps inside a trial and error loop. Trending because: 8 HuggingFace upvotes + trending in today’s HuggingFace papers feed.

13. The Next Screenshot Knows: Gated Hindsight Distillation for Mobile GUI Agents

Authors: Weiwei Li, Junzhuo Liu, Tong Chu, Hengfu Yu, Wen Li arXiv: arxiv.org/abs/2608.06065 Summary: GUI agents are commonly trained offline from successful interaction trajectories. Standard training decomposes each trajectory into prefix-action pairs: the agent predicts an action from the current screen and interaction history, while the subsequent observation is discarded. Trending because: 7 HuggingFace upvotes + trending in today’s HuggingFace papers feed.

14. Decoding-Level Taboo: A Diagnostic Stress Test for LLM Robustness

Authors: Tadanobu Chuyo Kamijo, Ori Rottenstreich, Javier Conde, Gonzalo Martínez, Pedro Reviriego arXiv: arxiv.org/abs/2608.09900 Summary: Large language model evaluations typically focus on performance under nominal conditions, creating an illusion of capability where models comfortably walk a narrow, highly optimized generation corridor. In real-world deployments, however, complex system prompts, safety guardrails, and structural constraints continuously force models off this nominal path, driving a divergence between benchmark scores and deployment performance. Trending because: 7 HuggingFace upvotes + trending in today’s HuggingFace papers feed.

15. WeClawArena: An Auditable Sandbox and Benchmark for Cross-User Agents Collaboration and Security in Human-Centered Agent Networks

Authors: Prince Zizhuang Wang, Aojie Yuan, Haiyue Zhang, Xiyang Hu, Yue Zhao, Shuli Jiang arXiv: arxiv.org/abs/2608.03499 Summary: Recent advances in persistent personal-agent frameworks are making human-centered agent networks realistic deployment targets: each user can be served by an AI agent that acts on the user’s behalf, maintains state, and communicates with other agents through social and task relations. In these networks, everyday tool use becomes multi-party owned-agent collaboration over personal workspaces, where files, records, tools, and policies are not directly visible across owners. Trending because: 6 HuggingFace upvotes + trending in today’s HuggingFace papers feed.

16. Omega-S: A Functional Resilience Index for LLM Fine-Tuning

Authors: Alberto Acedo arXiv: arxiv.org/abs/2608.03887 Summary: Fine-tuning a large language model on new data degrades what it previously learned. We present Omega-S, a drop-in penalty computed from the weight matrix alone: it needs no previous-task data, no Fisher matrix and no stored copy of the old weights. Trending because: 6 HuggingFace upvotes + trending in today’s HuggingFace papers feed.

17. The Loss Does Not See the Basis, but Adam Does

Authors: Devender Singh arXiv: arxiv.org/abs/2608.05136 Summary: Gradient descent on a factored model W = UV^top is implicitly biased toward low-rank solutions, while Adam, starting from the same small initialization, is not. We trace the difference to the gauge symmetry of the loss, its invariance under (U, V) mapsto (UQ, VQ). Trending because: 6 HuggingFace upvotes + trending in today’s HuggingFace papers feed.

18. Factorized Hypothesis Search for Evidence-to-Taxonomy Retrieval

Authors: Linhai Ma, Ethan F. Wei, Xueqing Peng, Yan Wang, Lingfei Qian, Víctor Gutiérrez-Basulto arXiv: arxiv.org/abs/2608.06614 Summary: Large-taxonomy retrieval often assumes that the input already expresses the target concept. In many settings, however, the input is indirect evidence, such as a table cell whose meaning depends on its row, column, datatype, and context. Trending because: 6 HuggingFace upvotes + trending in today’s HuggingFace papers feed.

19. Ego-OSCAR: Egocentric Open source Stereo CAptuRe System

Authors: Gunjan Paul, Senthil Palanisamy, Satpal Singh Rathore, Pratyush Kumar Patnaik, Shubhanshu Khatana, Abhishek Anand arXiv: arxiv.org/abs/2608.08285 Summary: We present Ego-OSCAR, an open-hardware, low-cost, head-mounted stereo-inertial capture device for egocentric data collection in the wild. EgoOSCAR pairs a hardware-synchronized global-shutter stereo camera with a 6- axis IMU, an embedded Linux SBC for on-device video encoding, and a realtime microcontroller for user feedback and watchdog functions. Trending because: 6 HuggingFace upvotes + trending in today’s HuggingFace papers feed.

20. VectraYX-Vision-1B: A Sub-2B Spanish/LATAM Cybersecurity Vision-Language Model with Structured Visual Reasoning and Native Tool Use

Authors: Juan S. Santillana arXiv: arxiv.org/abs/2608.08477 Summary: We present VectraYX-Vision-1B, a sub-2B vision-language model (VLM) for Spanish/LATAM cybersecurity imagery, coupling a frozen SigLIP-so400m encoder to a 1.04B Spanish/LATAM security decoder via an MLP. To our knowledge, it is the first sub-2B VLM specialized for cyber UI (IDA, Ghidra, Wireshark, Nmap, Metasploit, Volatility) that answers in Spanish, emits structured reasoning via native <|think|> tokens, invokes tools via Model Context Protocol (<|tool_call|>), and exports to llama.cpp’s LLaVA mmproj format for air-gapped deployment. Trending because: 6 HuggingFace upvotes + trending in today’s HuggingFace papers feed.