1
Ask HN: Temporal Awareness in LLM?
让大模型感知消息间的时间间隔，提升长篇叙事和日记的连贯性——ChatGPT用户提出的实用功能建议。
I have just sent a sort of "feature request" to ChatGPT (but my question is generally applicable to most LLMs). Here is the text, written by ChatGPT i…
2
Resource-Aware LLM Reasoning for Mobile Edge General Intelligence
移动边缘设备运行大模型？这篇论文提出了资源感知的LLM推理方案，让通用智能在边缘端高效落地。
arXiv:2509.23248v3 Announce Type: replace Abstract: The rapid advancement of large language models (LLMs) has enabled an emergence of agentic artifici…
3
PoQ-Judge: A Multi-Architecture Evaluation Framework for Cost-Aware Proof-of-Quality in Decentralized LLM Inference
去中心化LLM推理如何验证质量？新框架PoQ-Judge用多架构评估实现成本感知的质量证明。
arXiv:2606.11196v1 Announce Type: cross Abstract: Decentralized LLM inference networks need lightweight, reference-free quality evaluation for Proof o…
4
Carbon-Aware Governance Gates: An Architecture for Sustainable GenAI Development
提出碳感知治理门控架构，将碳排放纳入生成式AI开发决策，兼顾性能与可持续性。
arXiv:2602.19718v2 Announce Type: replace-cross Abstract: The rapid adoption of Generative AI (GenAI) in the software development life cycle (SDLC) in…
5
Conditional Vendi Score: Prompt-Aware Diversity Evaluation for Generative AI Models and LLMs
提出条件Vendi评分，首个能结合提示上下文评估生成模型多样性的方法，精准衡量AI创造力。
arXiv:2411.02817v2 Announce Type: replace-cross Abstract: Generative models guided by text prompts are widely evaluated for fidelity and prompt alignm…
6
ReflectiChain: Epistemic Grounding in LLM-Driven World Models for Supply Chain Resilience
论文提出ReflectiChain框架，通过认知接地弥补LLM在供应链世界模型中的物理感知缺失，提升韧性决策可靠性。
arXiv:2606.10359v1 Announce Type: new Abstract: AI agents in supply chains face a fundamental epistemic gap: large language models (LLMs) interpret po…
7
Quantifying Perception-Based Student Success with Generative AI: An Exploratory Monte Carlo Simulation
用蒙特卡洛模拟量化学生对生成式AI的感知成功，为教育评估提供新视角。
arXiv:2507.01062v4 Announce Type: replace-cross Abstract: Generative artificial intelligence (GenAI) tools such as ChatGPT have attracted growing atte…
8
SHAPE: Coalition-Aware Expert Pruning for Sparse Mixture-of-Experts LLMs
稀疏MoE大模型部署新突破：引入联盟感知策略的专家剪枝方法
arXiv:2606.09886v1 Announce Type: cross Abstract: Sparse Mixture-of-Experts (MoE) large language models achieve strong quality with low per-token comp…
9
POISE: Position-Aware Undetectable Skill Injection on LLM Agents
揭示LLM Agent技能注入新攻击，利用位置感知实现不可检测的恶意payload
arXiv:2606.07943v1 Announce Type: cross Abstract: Agent skills provide a lightweight mechanism for extending general-purpose agents, but their open fo…
10
Does Topic Sentiment Cause Perceived Ideology? Comparing Human and LLM Annotations in Political News Articles
探讨新闻话题情感是否影响人们对政治立场的感知，并对比人与LLM的标注差异，揭示大模型在意识形态推断上的潜力与局限。
arXiv:2606.06715v1 Announce Type: cross Abstract: We ask whether topic sentiment has a causal effect on perceived political ideology, and whether the …
11
Uncertainty-Aware LLM-Guided Policy Shaping for Sparse-Reward Reinforcement Learning
让大模型在稀疏奖励环境中引导强化学习策略，通过不确定性估计提升决策可靠性，有代码可复现。
arXiv:2606.06673v1 Announce Type: new Abstract: Sparse rewards and heterogeneous task sequences remain persistent challenges in Reinforcement Learning…
12
AAAC: Activation-Aware Adaptive Codebooks for 4-bit LLM Weight Quantization
提出AAAC方法，通过激活感知自适应码本，在保持4比特精度的同时进一步降低LLM权重量化误差
arXiv:2605.08692v2 Announce Type: replace Abstract: Post-training weight-only quantization to 4 bits is widely used to reduce the memory and compute c…
13
苹果 iOS 27 版 Siri 前瞻：AI 感知屏幕、跨 App 操作，新增独立应用
苹果iOS 27版Siri将迎来重大升级：AI屏幕感知、跨App操作、全新独立应用，果粉必看爆料！
IT之家 6 月 6 日消息，彭博社的马克 · 古尔曼（Mark Gurman）昨日（6 月 5 日）发表最新 Power On 时事通讯， 详细介绍了苹果 iOS 27 系统在 Siri 带来的诸多改变。 一、能力 能力方面，iOS 27 版 Siri 将围绕着用户画面（Personal Cont…
14
腾讯混元提出 Stem 稀疏注意力算法，首字延迟降低 3.6 倍
腾讯混元提出Stem稀疏注意力算法，从因果信息流革新块稀疏，首字延迟降低3.6倍，已被ICML-26收录。
IT之家 6 月 5 日消息，腾讯混元今日宣布提出 Stem 稀疏注意力算法，已被机器学习顶会 ICML-26 收录。 官方表示，Stem 稀疏注意力算法从“因果信息流”重新审视块级稀疏，用 Token 位置衰减（TPD）和输出感知度量（OAM）两大创新， 仅用 25% 算力就逼近稠密注意力的精度 …
15
Ask or Assume? Uncertainty-Aware Clarification-Seeking in Coding Agents
编程智能体何时该问、何时该猜？这篇论文提出不确定性感知的主动澄清策略，提升代码生成可靠性。
arXiv:2603.26233v2 Announce Type: replace Abstract: As Large Language Model (LLM) agents are increasingly deployed in open-ended domains like software…
16
Agentic Molecular Recovery via Molecule-Aware Exploration
新论文提出「分子感知探索」框架，让AI代理更智能地恢复和设计分子结构。
arXiv:2606.05847v1 Announce Type: new Abstract: Text-guided molecular generation with LLMs often yields invalid SMILES. We argue that invalid drafts s…
17
PAR3D: A Unified 3D-MLLM with Part-Aware Representation for Scene Understanding
突破性提出统一3D多模态大模型，部件感知表示让场景理解更细粒度。
arXiv:2606.06485v1 Announce Type: new Abstract: Recent advances in 3D multimodal large language models (3D-MLLMs) have enabled unified solutions for 3…
18
苹果专利构想 Vision Pro 配件：通过 Apple Pencil 触觉感知虚拟物体
IT之家 6 月 3 日消息，科技媒体 Appleinsider 昨日（6 月 2 日）发布博文，报道称苹果获批一项新专利，在配合 Vision Pro 头显使用时，Apple Pencil 通过触觉反馈传递虚拟物体的摩擦、纹理和粗糙感。 IT之家援引博文介绍，最新获批专利名称为《表面纹理检测与模拟…
19
DAPD: Dependency-Aware Parallel Decoding via Attention for Diffusion LLMs
扩散LLM迎来并行解码新突破，依赖感知注意力机制让加速生成更高效，ICML 2026接收
arXiv:2603.12996v2 Announce Type: replace Abstract: Parallel decoding for Diffusion LLMs (dLLMs) is difficult because each denoising step provides onl…
20
LaSR: Context-Aware Speech Recognition via Latent Reasoning
提出LaSR模型，用潜在推理机制提升语音识别的上下文感知能力，突破传统方法局限
arXiv:2606.00507v1 Announce Type: new Abstract: Recent advances in Speech Large Language Models (Speech LLMs) have significantly enhanced spoken langu…