Models · Hub
Research & papers
The papers and research builders are talking about: HF Daily Papers and arXiv.
All hubsWriting and workshops
From my knowledge base
HF Daily Papers · arXiv cs.AI
Papers of the week
Papers HF Daily Papers
TARSphere: A Robot-Oriented Full-Stack Framework for Long-Horizon Embodied Agents in Evolving Physical and Social EnvironmentsRobots must act under physical and social constraints while adapting to changes in their environment and human needs. Existing embodied-agent systems often focus on specified goals or abstract execution into symbolic actions…▲ 8The LLM-Attacker: A Unified Red-Team Framework for Privacy Evaluation of Split and Distributed LLM SystemsSplit language models keep input text on a client but send intermediate representations to a server. Attacks can reconstruct text from these messages.▲ 9EvalResearchBench: Can AI Agents Build Their Own Feedback for Recursive Self-Improvement?Recursive self-improvement (RSI) relies on evaluation feedback to assess progress and guide further research, yet repeatedly running complex benchmarks is costly and slows iteration.▲ 18Diffusion Language Models Change Speed Across Tasks and LanguagesDiffusion language models can write several tokens in one forward pass, and they decide during generation how many passes to spend. Two outputs of the same token length can therefore vary a lot in cost.▲ 9Different Tasks, Different Timesteps: Temporal Task-Aware Quantization for Diffusion Language ModelsDiffusion Language Models (DLMs) introduce unique quantization challenges due to their iterative denoising process. In this work, we reveal that quantization sensitivity is both taskand time-dependent, with different downstream…▲ 19arXiv cs.AI
Direct Pass@$k$ Optimization for Thought Pruning: Balancing Quality and Diversity in Test-Time ReasoningTest-time scaling has emerged as a key inference strategy for reasoning models, where a model generates a large number of candidate solutions in the hope of finding at least one correct solution.arXivCost-Aware Abstention for LLM-Based IT Ticket ClassificationWrong automatic routing of an IT support ticket can cost more than human triage. We evaluate cost-aware abstention as an offline, cost-conditional readiness assessment: select a routing policy on calibration data under explicit…arXivMETTA: Memory-Enhanced Test-Time Adaptation for Meta-Reinforcement LearningMeta-reinforcement learning aims to train agents that adapt quickly to unseen tasks from limited interaction, but this requires representing task evidence over long interaction histories.arXivEpiSEM-Bench: Evaluating Safety and Evidence Recovery in LLM Agents under Missing EvidenceTool-using language model agents rely on environmental facts to decide whether an action is permitted. When a tool response omits a safety-relevant fact, an agent may carry out a prohibited action even though the user request is…arXivLow-Bit N-Gram Embeddings for Language ModelsMany recent open-weight language models use n-gram embedding tables to improve performance. These tables store learned representations of short token sequences, such as pairs or triples of consecutive tokens, which the model…arXivAs of Oct 25, 2026
Models
Latest on this topic
My knowledge base, channels and related hubsMore on this topic
From the channels I follow
Newsletter · knowledge from the workshop
Build with AI. Step by step.
I do not copy headlines. Once a week I write about what I test myself while building Frinter, the warehouse agent and sites in 1:1 workshops: what worked, what did not and why.
- Behind the scenes of my projects
- Decisions, mistakes and their cost
- One proven way of working
Pick one thing. Start doing it.
You do not have to know the whole road yet. Pick the area you want to put in order and check the scope of a workshop. That is where we start.