Feed 订阅
7
个固定入口
Subscription View
趋势与订阅总览
从归档里提炼长期信号：每个 feed 的近 7 天 / 30 天命中情况，配置关键词的长期命中轨迹，以及跨多天反复出现的持续升温论文。
关键词追踪
65
个关键词页
最近 30 天
378
篇 feed 命中
持续升温论文
2
篇候选
阅读清单
2
篇已标记论文
Feed 趋势订阅
固定链接适合长期追踪某个研究方向的命中量和最近变化。
Feed
LM
LM 今日没有新的命中文献。
Feed
LLM
LLM 今日没有新的命中文献。
Feed
Agent Runtime Security
Agent Runtime Security 今日没有新的命中文献。
Feed
Vision
Vision 今日没有新的命中文献。
Feed
Terminal and SWE Agents
Terminal and SWE Agents 今日没有新的命中文献。
Feed
PubMed AI
PubMed AI 今日没有新的命中文献。
Feed
OpenAlex AI
《AI-Driven Multi-Agent System for Autonomous Mining Operation Centers》〔方法〕：International audience
关键词长期追踪
这些页面来自你配置里的 feed 关键词，适合观察某个主题是否持续冒头。
Topic
LLM
最近一次命中来自 LM：NuclearQAv2: A Structured Benchmark for Evaluating Domain-Science Competence in Large Language Models
Topic
language model
最近一次命中来自 LM：NuclearQAv2: A Structured Benchmark for Evaluating Domain-Science Competence in Large Language Models
Topic
benchmark
最近一次命中来自 LM：NuclearQAv2: A Structured Benchmark for Evaluating Domain-Science Competence in Large Language Models
Topic
large language model
最近一次命中来自 LM：NuclearQAv2: A Structured Benchmark for Evaluating Domain-Science Competence in Large Language Models
Topic
agent
最近一次命中来自 LM：Joint Learning of Experiential Rules and Policies for Large Language Model Agents
Topic
evaluation
最近一次命中来自 LM：NuclearQAv2: A Structured Benchmark for Evaluating Domain-Science Competence in Large Language Models
Topic
reasoning
最近一次命中来自 LM：NuclearQAv2: A Structured Benchmark for Evaluating Domain-Science Competence in Large Language Models
Topic
RAG
最近一次命中来自 LM：NuclearQAv2: A Structured Benchmark for Evaluating Domain-Science Competence in Large Language Models
Topic
alignment
最近一次命中来自 LM：In-Context Model Predictive Generation: Open-Vocabulary Motion Synthesis from Language Models to Physics
Topic
coding agent
最近一次命中来自 Terminal and SWE Agents：A Deterministic Control Plane for LLM Coding Agents
Topic
jailbreak
最近一次命中来自 Agent Runtime Security：Jailbreaking for the Average Jane: Choosing Optimal Jailbreaks via Bandit Algorithms for Automatically Enhanced Queries
Topic
guardrail
最近一次命中来自 Agent Runtime Security：Do Safety Guardrails Need to Reason? LeanGuard: A Fast and Light Approach for Robust Moderation
Topic
prompt injection
最近一次命中来自 LM：Prompt Injection in Automated Résumé Screening with Large Language Models: Single and Multi-Injection Settings
Topic
SWE-bench
最近一次命中来自 Terminal and SWE Agents：To Run or Not to Run: Analyzing the Cost-Effectiveness of Code Execution in LLM-Based Program Repair
Topic
computer-use agent
最近一次命中来自 LM：Uncertainty Quantification for Computer-Use Agents: A Benchmark across Vision-Language Models and GUI Grounding Datasets
Topic
Terminal-Bench
最近一次命中来自 Terminal and SWE Agents：LemonHarness Technical Report
Topic
code agent
最近一次命中来自 Terminal and SWE Agents：How Much Static Structure Do Code Agents Need? A Study of Deterministic Anchoring
Topic
instruction tuning
最近一次命中来自 LM：SARA: Unlocking Multilingual Knowledge in Mixture-of-Experts via Semantically Anchored Routing Alignment
Topic
repository-level
最近一次命中来自 Terminal and SWE Agents：Evaluating LLMs on Real-World Software Performance Optimization
Topic
in-context learning
最近一次命中来自 LM：MedGuards: Multi-Agent System for Reliable Medical Error Detection and Correction
Topic
agent runtime
最近一次命中来自 Agent Runtime Security：Getting Better at Working With You: Compiling User Corrections into Runtime Enforcement for Coding Agents
Topic
indirect prompt injection
最近一次命中来自 Agent Runtime Security：CodeSentinel: A Three-Layer Defense Against Indirect Prompt Injection in Code Contexts
Topic
policy enforcement
最近一次命中来自 LM：A Technical Taxonomy of LLM Agent Communication Protocols
Topic
code editing
最近一次命中来自 Agent Runtime Security：WeaveBench: A Long-Horizon, Real-World Benchmark for Computer-Use Agents with Hybrid Interfaces
Topic
code generation benchmark
最近一次命中来自 Terminal and SWE Agents：VoidPadding: Let [VOID] Handle Padding in Masked Diffusion Language Models so that [EOS] Can Focus on Semantic Termination
Topic
issue resolution
最近一次命中来自 Terminal and SWE Agents：Unlocking Model Potentials Through Adaptive Multi-Agent Scaffolding for Efficient Issue Resolution
Topic
program repair
最近一次命中来自 Terminal and SWE Agents：Smaller Models, Unexpected Costs: Trade-offs in LLM Quantization for Automated Program Repair
Topic
sandboxing
最近一次命中来自 Agent Runtime Security：Burnyard: Future of Malware Analysis
Topic
agent security
最近一次命中来自 Agent Runtime Security：Toward Secure LLM Agents: Threat Surfaces, Attacks, Defenses, and Evaluation
Topic
automated program repair
最近一次命中来自 Terminal and SWE Agents：Smaller Models, Unexpected Costs: Trade-offs in LLM Quantization for Automated Program Repair
Topic
data exfiltration
最近一次命中来自 Agent Runtime Security：Securing LLM-Agent Long-Term Memory Against Poisoning: Non-Malleable, Origin-Bound Authority with Machine-Checked Guarantees
Topic
secure agent
最近一次命中来自 Agent Runtime Security：Provably Secure Agent Guardrail
Topic
SWE bench
最近一次命中来自 Terminal and SWE Agents：Exploration Structure in LLM Agents for Multi-File Change Localization
Topic
terminal agent
最近一次命中来自 Terminal and SWE Agents：Tmax: A simple recipe for terminal agents
Topic
test generation
最近一次命中来自 Terminal and SWE Agents：Knowledge Matters: Injecting Project and Testing Knowledge into LLM-based Unit Test Generation
Topic
privilege escalation
最近一次命中来自 Agent Runtime Security：Seeing Is Not Screening: Multimodal Hidden Instruction Attacks on Agent Skill Scanners
Topic
retrieval augmented generation
最近一次命中来自 LM：Probabilistic Agents in Deterministic Audits: Evaluating Multi-Agent Systems for Automated Audits Based on the German IT-Grundschutz
Topic
agent attack
最近一次命中来自 Agent Runtime Security：Secure UAV Swarms in Low-Altitude Wireless Networks: Challenges and Solutions
Topic
agent defense
最近一次命中来自 Agent Runtime Security：SafeMCP: Proactive Power Regulation for LLM Agent Defense via Environment-Grounded Look-Ahead Reasoning
Topic
agent isolation
最近一次命中来自 LLM：RPA-Check: A Multi-Stage Automated Framework for Evaluating Dynamic LLM-based Role-Playing Agents
Topic
agent sandbox
最近一次命中来自 Agent Runtime Security：DeltaBox: Scaling Stateful AI Agents with Millisecond-Level Sandbox Checkpoint/Rollback
Topic
bug fixing
最近一次命中来自 Terminal and SWE Agents：DeNovoSWE: Scaling Long-Horizon Environments for Generating Entire Repositories from Scratch
Topic
code repair
最近一次命中来自 Terminal and SWE Agents：SHERLOC: Structured Diagnostic Localization for Code Repair Agents
Topic
LLM agent security
最近一次命中来自 Agent Runtime Security：Toward Secure LLM Agents: Threat Surfaces, Attacks, Defenses, and Evaluation
Topic
malicious tool
最近一次命中来自 Agent Runtime Security：From Control Boundary to Insurance Claim: Reconstructing AI-Mediated Losses Through the CER Framework
Topic
multimodal language model
最近一次命中来自 LLM：Don't Show Pixels, Show Cues: Unlocking Visual Tool Reasoning in Language Models via Perception Programs
Topic
patch generation
最近一次命中来自 Agent Runtime Security：EviACT: An Evidence-to-Action Framework for Agentic Program Repair
Topic
repository level
最近一次命中来自 Terminal and SWE Agents：Dependency-Guided Repository-Level C-to-Rust Translation with Reinforcement Alignment
Topic
runtime security
最近一次命中来自 LLM：ClawGuard: A Runtime Security Framework for Tool-Augmented LLM Agents Against Indirect Prompt Injection
Topic
shell agent
最近一次命中来自 Agent Runtime Security：How Agentic AI Coding Assistants Become the Attacker's Shell
Topic
software engineering agent
最近一次命中来自 Terminal and SWE Agents：Same Signal, Different Semantics: A Cross-Framework Behavioral Analysis of Software Engineering Agents
Topic
terminal bench
最近一次命中来自 LM：What Makes a Good Terminal-Agent Benchmark Task: A Guideline for Adversarial, Difficult, and Legible Evaluation Design
Topic
AI agent security
暂未命中，页面会持续追踪后续归档。
Topic
autonomous agent security
暂未命中，页面会持续追踪后续归档。
Topic
browser agent security
暂未命中，页面会持续追踪后续归档。
Topic
code agent security
暂未命中，页面会持续追踪后续归档。
Topic
command line agent
暂未命中，页面会持续追踪后续归档。
Topic
function calling security
暂未命中，页面会持续追踪后续归档。
Topic
MCP security
暂未命中，页面会持续追踪后续归档。
Topic
model context protocol security
暂未命中，页面会持续追踪后续归档。
Topic
software engineering benchmark
暂未命中，页面会持续追踪后续归档。
Topic
terminal benchmark
暂未命中，页面会持续追踪后续归档。
Topic
tool calling security
暂未命中，页面会持续追踪后续归档。
Topic
tool-use security
暂未命中，页面会持续追踪后续归档。
Topic
untrusted tool
暂未命中，页面会持续追踪后续归档。
持续升温论文
这些论文在多个日期或多个 feed 中反复出现，更适合放进长期观察列表。
Momentum
Lightweight Multimodal Adaptation of Vision Language Models for Species Recognition and Habitat Context Interpretation in Drone Thermal Imagery
This study proposes a lightweight multimodal adaptation framework to bridge the representation gap between RGB-pretrained VLMs and thermal infrared imagery, and demonstrates its practical utility using a real drone-coll…
Momentum
MMEmb-R1: Reasoning-Enhanced Multimodal Embedding with Pair-Aware Selection and Adaptive Control
MLLMs have been successfully applied to multimodal embedding tasks, yet their generative reasoning capabilities remain underutilized. Directly incorporating chain-of-thought reasoning into embedding learning introduces…