Unzip

## Episode Summary

In this episode, we cover:

- **Benchmark Test-Time Scaling of General LLM Agents** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.18998)

- **Learning from Trials and Errors: Reflective Test-Time Planning for Embodied LLMs** (arXiv)

- [Read more](http://arxiv.org/abs/2602.21198v1)

- **On Data Engineering for Scaling LLM Terminal Capabilities** (arXiv)

- [Read more](http://arxiv.org/abs/2602.21193v1)

- **See and Fix the Flaws: Enabling VLMs and Diffusion Models to Comprehend Visual Artifacts via Agentic Data Synthesis** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.20951)

- **Communication-Inspired Tokenization for Structured Image Representations** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.20731)

---

*Sponsored by LimitLess AI*

## Episode Summary

In this episode, we cover:

- **Anatomy of Agentic Memory: Taxonomy and Empirical Analysis of Evaluation and System Limitations** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.19320)

- **Skill-Inject: Measuring Agent Vulnerability to Skill File Attacks** (arXiv)

- [Read more](http://arxiv.org/abs/2602.20156v1)

- **To Reason or Not to: Selective Chain-of-Thought in Medical Question Answering** (arXiv)

- [Read more](http://arxiv.org/abs/2602.20130v1)

- **Decoding ML Decision: An Agentic Reasoning Framework for Large-Scale Ranking System** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.18640)

- **Nacrith: Neural Lossless Compression via Ensemble Context Modeling and High-Precision CDF Coding** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.19626)

---

*Sponsored by LimitLess AI*

## Episode Summary

In this episode, we cover:

- **ReIn: Conversational Error Recovery with Reasoning Inception** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.17022)

- **VESPO: Variational Sequence-Level Soft Policy Optimization for Stable Off-Policy LLM Training** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.10693)

- **Does Your Reasoning Model Implicitly Know When to Stop Thinking?** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.08354)

- **VIRAASAT: Traversing Novel Paths for Indian Cultural Reasoning** (arXiv)

- [Read more](http://arxiv.org/abs/2602.18429v1)

- **Rubrics as an Attack Surface: Stealthy Preference Drift in LLM Judges** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.13576)

---

*Sponsored by LimitLess AI*

## Episode Summary

In this episode, we cover:

- **"What Are You Doing?": Effects of Intermediate Feedback from Agentic LLM In-Car Assistants During Multi-Step Processing** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.15569)

- **Mobile-Agent-v3.5: Multi-platform Fundamental GUI Agents** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.16855)

- **Calibrate-Then-Act: Cost-Aware Exploration in LLM Agents** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.16699)

- **World Models for Policy Refinement in StarCraft II** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.14857)

- **OpenEarthAgent: A Unified Framework for Tool-Augmented Geospatial Agents** (arXiv)

- [Read more](http://arxiv.org/abs/2602.17665v1)

---

*Sponsored by LimitLess AI*

## Episode Summary

In this episode, we cover:

- **"What Are You Doing?": Effects of Intermediate Feedback from Agentic LLM In-Car Assistants During Multi-Step Processing** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.15569)

- **Mobile-Agent-v3.5: Multi-platform Fundamental GUI Agents** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.16855)

- **Calibrate-Then-Act: Cost-Aware Exploration in LLM Agents** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.16699)

- **World Models for Policy Refinement in StarCraft II** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.14857)

- **OpenEarthAgent: A Unified Framework for Tool-Augmented Geospatial Agents** (arXiv)

- [Read more](http://arxiv.org/abs/2602.17665v1)

---

*Sponsored by LimitLess AI*

## Episode Summary

In this episode, we cover:

- **"What Are You Doing?": Effects of Intermediate Feedback from Agentic LLM In-Car Assistants During Multi-Step Processing** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.15569)

- **Mobile-Agent-v3.5: Multi-platform Fundamental GUI Agents** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.16855)

- **Calibrate-Then-Act: Cost-Aware Exploration in LLM Agents** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.16699)

- **World Models for Policy Refinement in StarCraft II** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.14857)

- **OpenEarthAgent: A Unified Framework for Tool-Augmented Geospatial Agents** (arXiv)

- [Read more](http://arxiv.org/abs/2602.17665v1)

---

*Sponsored by LimitLess AI*

## Episode Summary

In this episode, we cover:

- **BiManiBench: A Hierarchical Benchmark for Evaluating Bimanual Coordination of Multimodal Large Language Models** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.08392)

- **Learning Situated Awareness in the Real World** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.16682)

- **MAEB: Massive Audio Embedding Benchmark** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.16008)

- **RynnBrain: Open Embodied Foundation Models** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.14979)

- **Towards a Science of AI Agent Reliability** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.16666)

---

*Sponsored by LimitLess AI*

## Episode Summary

In this episode, we cover:

- **SkillsBench: Benchmarking How Well Agent Skills Work Across Diverse Tasks** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.12670)

- **How Much Reasoning Do Retrieval-Augmented Models Add beyond LLMs? A Benchmarking Framework for Multi-Hop Inference over Hybrid Knowledge** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.10210)

- **The Vision Wormhole: Latent-Space Communication in Heterogeneous Multi-Agent Systems** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.15382)

- **A Trajectory-Based Safety Audit of Clawdbot (OpenClaw)** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.14364)

- **TAROT: Test-driven and Capability-adaptive Curriculum Reinforcement Fine-tuning for Code Generation with Large Language Models** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.15449)

---

*Sponsored by LimitLess AI*

## Episode Summary

In this episode, we cover:

- **VisPhyWorld: Probing Physical Reasoning via Code-Driven Video Reconstruction** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.13294)

- **DeepImageSearch: Benchmarking Multimodal Agents for Context-Aware Image Retrieval in Visual Histories** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.10809)

- **CellMaster: Collaborative Cell Type Annotation in Single-Cell Analysis** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.13346)

- **Nanbeige4.1-3B: A Small General Model that Reasons, Aligns, and Acts** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.13367)

- **Learning to Configure Agentic AI Systems** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.11574)

---

*Sponsored by LimitLess AI*

## Episode Summary

In this episode, we cover:

- **SQuTR: A Robustness Benchmark for Spoken Query to Text Retrieval under Acoustic Noise** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.12783)

- **Less is Enough: Synthesizing Diverse Data in Feature Space of LLMs** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.10388)

- **What does RL improve for Visual Reasoning? A Frankenstein-Style Analysis** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.12395)

- **SemanticMoments: Training-Free Motion Similarity via Third Moment Features** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.09146)

- **Best of Both Worlds: Multimodal Reasoning and Generation via Unified Discrete Flow Matching** (Hugging Face Daily)

- [Read more](https://huggingface.co/papers/2602.12221)

---

*Sponsored by LimitLess AI*