📑 前沿论文

Paper Readings

Deep dives into the latest AI, ML, and systems research papers.

🔍
2026-11-21
AI Papers Weekly: Agentic Robotics, Open MoE Fleets, and Hardware-Aware Systems

Explore the shift to agentic robotics, K2-Horizon's open MoE models, and hardware-specific LLM optimization. Key insights for AI engineers on agent security, evaluation benchmarks, and systems engineering.

#papers#weekly-papers#ai-research#huggingface#agentic-ai#llm-infrastructure#robotics
2026-11-20
AI Papers Daily: Agentic Forensics, Embodied Benchmarks, and Code Search Realities

Nov 20, 2026: Deep dive into WeVisForensics for video detection, Mind2Act World for robotics, and Repo-Premium for code search. Key insights for AI engineers on agentic reliability and evaluation.

#papers#daily-papers#ai-research#huggingface#agentic-ai#evaluation-benchmarks#code-search
2026-11-19
AI Papers Daily: Agent Security, Quantization, and System One Generation

Explore PIPETTE for agent security, QDA for automated quantization, and Jeveration for single-pass image generation. Key insights for AI engineers on robustness and efficiency.

#papers#daily-papers#ai-research#huggingface#llm-agents#mlops#multimodal
2026-11-18
AI Papers Daily: Agentic Infrastructure, Kernel Optimization, and Multimodal Reasoning

Explore Xing4's agentic training on domestic chips, AscendGYM's kernel optimization benchmark, and new approaches to multimodal reasoning and coding agents.

#papers#daily-papers#ai-research#huggingface#llm-agents#mlops#multimodal
2026-11-17
AI Papers Daily: K2-Horizon Open Models, SWE-CoWork Agents, and Inclusive Med AI

Explore K2-Horizon's open MoE fleet, SWE-CoWork's organizational coding benchmarks, and InclusionMed's healthcare AI evaluation framework for AI engineers.

#papers#daily-papers#ai-research#huggingface#llm-agents#moe-architecture#benchmarking
2026-11-16
AI Papers Daily: Agentic Robotics, Web Agents, and Memory Systems

Explore agentic robotics, web agent benchmarks, and multi-agent memory systems. Key insights for AI engineers on LLM agents, evaluation, and inference optimization.

#papers#daily-papers#ai-research#huggingface#llm-agents#inference-optimization#robotics
2026-11-14
AI Papers Weekly: Agentic Efficiency, Embodied Intelligence, and Inference Optimization

Discover key papers on agentic model collaboration, embodied robotics, and efficient LLM inference. Learn how to optimize agent costs, handle multimodal data, and design reliable AI systems for your next interview.

#papers#weekly-papers#ai-research#huggingface#agentic-ai#embodied-intelligence#inference-optimization
2026-11-13
AI Papers Daily: System One Benchmarks, Productivity Agents, and Open Vision Recipes

Nov 13, 2026: SO-Arena benchmarks fast decision-making, Agent-R^2 scales productivity workflows, and Open Vision Banana offers a fully open generative perception recipe.

#papers#daily-papers#ai-research#huggingface#llm-agents#inference-optimization#multimodal
2026-11-12
AI Papers Daily: Agent Quantization, Adaptive Serving, and Self-Learning RL

Explore SteerDistill for quantized agent recovery, Evolving Infra for adaptive LLM serving, and RL-driven self-learning. Key insights for AI engineers on inference efficiency and agent evaluation.

#papers#daily-papers#ai-research#huggingface#llm-agent#inference-optimization#rlhf
2026-11-11
AI Papers Daily: Bit-Level LLM Repairs, Agent Memory Safety, and Agentic Robotics

Explore Tree-Ising for cheap LLM weight repairs, WikiTrace for safer agent memory, and new benchmarks for agentic robotics and finance. Key insights for AI engineers.

#papers#daily-papers#ai-research#huggingface#llm-agents#model-repair#embodied-ai