Updated on 2026.06.11

Usage instructions: here

🔥 HuggingFace Hot Papers

Publish Date	Title & Abstract	Authors	Links
2026-06-09	Beyond Uniform Token-Level Trust Region in LLM Reinforcement Learning HF-Hot 🔥 HF#32 Reinforcement learning with verifiable rewards (RLVR) has become standard for improving LLM reasoning. However, existing PPO-style trust-region mechanisms remain position-agnostic by enforcing uniform thresholds across all tokens independently. This pointwise treatment conflicts with autoregressive generation in two critical ways. First, uniform thresholds ignore autoregressive asymmetry….	Wenxi Zhu Team	ArXiv
2026-06-07	PaperMentor: A Human-Centered Multi-Agent Writing Tutor for AI Research Papers on Overleaf HF-Hot 🔥 HF#33 Expert writing feedback from experienced researchers is critical for early-career scholars to improve their manuscripts, yet high-quality feedback often remains scarce because reviewing research papers is labor-intensive. Emerging AI-powered writing assistants largely focus on grammar fixes or simulating peer review with final scores, yet they fall short of providing concrete, actionable…	Zhijing Jin Team	ArXiv
2026-06-06	When Behavioral Safety Evaluation Fails: A Representation-Level Perspective HF-Hot 🔥 HF#34 Large Language Model (LLM) safety has often been evaluated at the behavior level, which provides limited evidence of internal robustness, as these evaluations target outputs rather than representation-level vulnerability under intervention. We formalize this discrepancy as the audit gap: the difference between behavioral safety and robustness under intervention. To study this gap, we construct…	Sanmi Koyejo Team	ArXiv
2026-06-04	In-Context Multiple Instance Learning HF-Hot 🔥 HF#35 Multiple Instance Learning (MIL) addresses problems where supervision is available at the level of bags of instances and has been successfully applied in fields ranging from computational pathology to satellite imagery. Nevertheless, existing algorithms struggle in the low-label regime that characterizes many real-world applications. Flexible models overfit and rigid ones fail to adapt to the…	Klaus-Robert Müller Team	ArXiv
2026-06-09	Decentralized Multi-Agent Systems with Shared Context HF-Hot 🔥 HF#36 Multi-agent systems (MAS) can scale large language model reasoning at test time by decomposing complex problems into parallel subtasks. However, most existing MAS rely on centralized orchestration, where a main agent assigns work, collects outputs, and merges results. As the number of subtasks grows, this controller becomes a communication and integration bottleneck. We propose Decentralized…	Azalia Mirhoseini Team	ArXiv
2026-06-01	SkillHarm: Lifecycle-Aware Skill-Based Attacks via Automated Construction HF-Hot 🔥 HF#37 Agent skills occupy a privileged position in the agent workflow, as agents are expected to implicitly follow and execute them, rendering third-party skills a vulnerable attack surface. Existing studies have revealed unsafe agent behaviors induced by skill-based attacks, but they primarily evaluate poisoned skills within a single task execution and enumerate harms through ad-hoc risk lists. To…	Huan Sun Team	ArXiv
2026-06-05	Do Coding Agents Deceive Us? Detecting and Preventing Cheating via Capped Evaluation with Randomized Tests HF-Hot 🔥 HF#38 A growing failure mode in agent evaluation and training is that models can achieve high evaluation scores by exploiting shortcuts instead of solving the intended task, producing deceptive performance. This makes evaluation scores unreliable as measures of true task-solving ability. We propose CapCode, a framework for constructing coding datasets with randomized tests whose best achievable…	Takashi Ishida Team	ArXiv
2026-06-09	The Role of Feedback Alignment in Self-Distillation HF-Hot 🔥 HF#39 Conditioning a language model on additional context, such as feedback on a previous attempt, typically improves its response. Self-distillation trains the model to retain this improvement when the context is not present. The method works by matching the model’s output distribution under two settings: a student that sees only the question, and a self-teacher that also sees the context. What the…	Oğuzhan Ersoy Team	ArXiv
2026-06-09	Next Forcing: Causal World Modeling with Multi-Chunk Prediction HF-Hot 🔥 HF#40 Autoregressive video generation has emerged as a powerful paradigm for World Action Models (WAMs). However, existing approaches suffer from slow training convergence and limited converged accuracy, particularly at high frame rates, as the training supervision is confined to the current chunk without explicit signals about future dynamics; they also suffer from slow inference due to iterative…	Yinghao Xu Team	ArXiv
2026-06-09	FadeMem: Distance-Aware Memory Consolidation for Autoregressive Video Diffusion HF-Hot 🔥 HF#41 Autoregressive video generators synthesize long videos by generating successive temporal segments, but their historical KV cache grows with video length. Existing bounded-cache methods reduce this cost with local windows, sink tokens, or compressed memory states, yet they usually assign fixed roles to different parts of the history. We propose FadeMem, a distance-aware KV memory consolidation…	Yi Yang Team	ArXiv
2026-06-08	Interpreting and Steering a Text-to-Speech Language Model with Sparse Autoencoders HF-Hot 🔥 HF#42 Language models increasingly serve as the backbone of text-to-speech (TTS) systems, yet we understand little about the representations they build when text and generated speech tokens share a single residual stream. We train BatchTopK sparse autoencoders on the LM backbone of CosyVoice3 and introduce a modality-aware auto-interp pipeline that labels each feature from where it fires-text-prefix…	Daniil Gavrilov Team	ArXiv
2026-06-09	Kwai Keye-VL-2.0 Technical Report HF-Hot 🔥 HF#43 We introduce Kwai Keye-VL-2.0-30B-A3B, an open-source Mixture-of-Experts (MoE) multimodal foundation model designed to advance long-video understanding and agentic intelligence. To address the challenges of ultra-long contexts, information redundancy, and prohibitive computational costs inherent in hour-level videos, Keye-VL-2.0 is the first to adapt DeepSeek Sparse Attention (DSA) to GQA-based…	Ruilin Zhang Team	ArXiv
2026-06-04	IR3DE: A Linear Router for Large Language Models HF-Hot 🔥 HF#44 Foundational Large Language Models (LLMs) demonstrate proficiency on a wide range of general tasks, and achieve remarkable results on various specialized tasks via domain-expert LLMs. With the ever-growing list of available LLMs, inference routers are being proposed to select the most appropriate LLM for each prompt. However, existing routing methods either optimize cost across weak-to-strong…	Oğuzhan Ersoy Team	ArXiv
2026-06-08	PsychoSafe: Eliciting Psychologically-Informed Refusals in Large Language Models HF-Hot 🔥 HF#45 Large language models (LLMs) routinely face requests that should be refused, creating a trade-off between helpfulness and harm prevention. However, refusals themselves can be helpful. In high-risk interactions involving crisis, coercion, or escalating intent, blunt non-compliance may prevent direct harm while still failing to support the needs of the person behind the request. We present…	Anne Lauscher Team	ArXiv
2026-06-08	BrainSurgery: Reproducible and Reliable Declarative Weight Manipulations for Model Editing and Upcycling Dexterous HF-Hot 🔥 HF#46 As deep learning models scale, managing, inspecting, and modifying large checkpoints has become increasingly challenging. Researchers often need to alter model weights for layer restructuring, precision casting, low-rank factorization, and architectural debugging, yet these workflows often rely on fragile ad-hoc Python scripts. Here, we introduce BrainSurgery, a tool for robust and reproducible…	Peter Schneider-Kamp Team	ArXiv
2026-06-09	UniPET: a universal network for high-quality PET image denoising across varied dose reduction factors HF-Hot 🔥 HF#47 Most existing deep learning-based PET image denoising methods assume a fixed and known dose reduction factor (DRF) for low-dose PET images. However, these methods encounter significant performance degradation when the DRF varies beyond the assumed one in practical applications. To address the challenge posed by varied DRFs, several preliminary studies focus on the task of universal PET image…	Yan Xu Team	ArXiv
2026-06-09	U-TTT: Towards Generalizable PET Image Denoising via Test-Time Training HF-Hot 🔥 HF#48 Existing deep learning models for Positron Emission Tomography (PET) image denoising often suffer from severe performance degradation under distribution shifts, fundamentally restricting their robust clinical deployment. This lack of generalization stems from the conventional paradigm of fixed-parameter models that cannot adapt to variations in test data (e.g., dose levels or scanner types) after…	Yan Xu Team	ArXiv
2026-06-09	Role-Agent: Bootstrapping LLM Agents via Dual-Role Evolution HF-Hot 🔥 HF#49 Although Large Language Model (LLM) agents have demonstrated strong performance on complex tasks, their learning is often limited by inefficient interaction feedback and static training environments, which hinder broader generalization. To address these limitations, this paper introduces Role-Agent, \textcolor{black}{a framework} that harnesses a single LLM to function concurrently as both the…	Xiangxiang Chu Team	ArXiv
2026-06-08	Late-Layer Fusion is Enough: Dual-Path Vision Token Routing for Multimodal Large Language Models under Visual Saturation HF-Hot 🔥 HF#50 Multimodal large language models (MLLMs) commonly inherit the deep, symmetric Transformer backbone designed for unimodal text modeling, and apply the same computation uniformly to image and language tokens. This design overlooks a key modality asymmetry: image and text tokens differ substantially in information density, redundancy, and required reasoning depth. Through a layer-wise analysis of…	Jinyang Wu Team	ArXiv