Xeno-Interpretability: Investigating the Alien Minds of LLMs
arXiv:2609.20408v2 Announce Type: replace-cross Abstract: Large language models are usually interpreted through concepts that humans already possess: truthfulness, refusal, deception,…
Didactic knowledge or Clinical Cases? How Data Types Shape Medical Large Language Models
arXiv:2609.22161v1 Announce Type: new Abstract: Medical large language models are commonly trained on mixtures of didactic data (e.g., textbooks) and…
GameLogicBench: Evaluating Coding Agents on Runtime Game Logic with Tick-Level State Assertions
arXiv:2609.21562v2 Announce Type: replace-cross Abstract: Coding agents can modify and test code across large software projects. Game development is a…
InterHier: Learning Interconnected Hierarchical Semantics for Open-Vocabulary Object Detection
arXiv:2609.24026v1 Announce Type: cross Abstract: In this paper, we investigate the limitations of fixed, hand-crafted connectors in hierarchical semantic representations…
Misaligned Clinical Risk Classification and Cost Asymmetry in Open-Weight Large Language Models
arXiv:2609.23999v1 Announce Type: cross Abstract: How large language models (LLMs) integrate patient risk with clinical cost tradeoffs remains poorly understood.…
Paint-Anything: Unified Any-Color Control for Image Generation and Editing
arXiv:2609.20816v2 Announce Type: replace-cross Abstract: Professional design requires any-color control: the ability to specify an object’s target color with any…
RBS-Attention: Radius-Bounded Sparse Prefill for Long-Context Large Language Models
arXiv:2609.20971v1 Announce Type: new Abstract: Long-context large language model inference is increasingly limited by prefill, where dense self-attention processes the…
PACE: Precise AI Cinematic Expression
arXiv:2609.19853v2 Announce Type: replace-cross Abstract: Between a screenplay and a film sits a planning problem that is spatial first: who…
TERMon: Detecting Persistent Behavioral Threats in Edge AI via Hardware-Native Ternary Runtime Monitor
arXiv:2609.21713v1 Announce Type: cross Abstract: Edge AI accelerators are increasingly deployed in safety-critical environments, where model outputs may control physical…
CIPL: A Channel-Aware Framework for Recoverable Privacy Leakage in LLM Agents
arXiv:2609.21686v1 Announce Type: cross Abstract: Privacy leakage in LLM agents is commonly evaluated within individual components such as memory, retrieval,…