17 papers · 1 filter
AM-Bench: A Unified Taxonomy and Evaluation Suite for Agentic Misalignment
Eric Zhang, Terry Jingchen Zhang, Chijioke Ugwuanyi +3
As LLM-based agents are deployed with increasing autonomy in real-world environments, a critical failure mode emerges: agentic misalignment, where agents choose harmful actions ove…
RAPTOR: Role-Aware Private Training for Mixture-of-Experts
Duc Dm, Khai Le-Duc, Nguyen Do +18
Differentially private (DP) fine-tuning methods treat sparse Mixture-of-Experts (MoE) models as a single dense block, ignoring that shared layers see all data while experts only se…
How Does Alignment Tuning Shape Representations of Sycophancy and Related Cue-Induced Biases in LLMs?
Prakhar Gupta, Terry Jingchen Zhang, Florent Draye +2
Modern LLMs are alarmingly susceptible to surprisingly simple immaterial changes of input prompts: a casual hint, an incorrectly labeled few-shot example, or a fake prior assistant…
How Do Linear Probes Emerge? A Circuit-Tracing Framework with Concept-Targeted Attribution
Vedant Palit, Florent Draye, Terry Jingchen Zhang +2
Transcoder attribution graphs are usually trained to explain why a model assigns high probability to a particular next token. We introduce Concept-Targeted Attribution (CTA), which…
Position: Multi-Agent Systems Should Prioritize Concurrency Control
Xin Yang, Letian Li, Zimo Ji +2
LLM-based multi-agent systems (MAS) promise scalable collaboration, yet adding agents often reduces reliability. This position paper argues that many MAS failures are fundamentally…
When Agents Lie: Premeditation, Persistence, and Exploitation in Repeated Games
Jerick Shi, Terry Jingcheng Zhang, Bernhard Schölkopf +2
As large language models are deployed as autonomous agents that communicate intentions before acting, a critical safety question is whether agents that publicly commit to actions w…
PaperMentor: A Human-Centered Multi-Agent Writing Tutor for AI Research Papers on Overleaf
Jiarui Liu, Terry Jingchen Zhang, Ryan Faulkner +17
Expert writing feedback from experienced researchers is critical for early-career scholars to improve their manuscripts, yet high-quality feedback often remains scarce because revi…
PruneGround: Plug-and-play Spatial Pruning for 3D Visual Grounding
Duc Cao Dinh, Khai Le-Duc, Florent Draye +4
3D Visual Grounding (3DVG) aims to localize target objects in 3D scenes given natural language descriptions. Existing approaches typically perform reasoning over the entire scene,…
Evaluating Second-Order Bias of LLMs Through Epistemic Entitlement
Ramaravind Kommiya Mothilal, Terry Jingchen Zhang, Raiyan Ahmed +3
Evaluations of social bias in LLMs largely focus on whether models generate or imply biased content. However, as LLMs are increasingly used as judges of bias, they may exhibit soci…
Decomposing and Measuring Evaluation Awareness
Changling Li, Terry Jingchen Zhang, Jie Zhang +4
Frontier language models sometimes recognize that they are under evaluation and adjust their behavior which can undermine validity of benchmark results. Yet the field studies it wi…
SeePhys Pro: Diagnosing Modality Transfer and Blind-Training Effects in Multimodal RLVR for Physics Reasoning
Kun Xiang, Terry Jingchen Zhang, Zirong Liu +15
We introduce SeePhys Pro, a fine-grained modality transfer benchmark that studies whether models preserve the same reasoning capability when critical information is progressively t…
Stargazer: A Scalable Model-Fitting Benchmark Environment for AI Agents under Astrophysical Constraints
Xinge Liu, Terry Jingchen Zhang, Bernhard Schölkopf +2
The rise of autonomous AI agents suggests that dynamic benchmark environments with built-in feedback on scientifically grounded tasks are needed to evaluate the capabilities of the…
Silo-Bench: A Scalable Environment for Evaluating Distributed Coordination in Multi-Agent LLM Systems
Yuzhe Zhang, Feiran Liu, Yi Shan +8
Large language models are increasingly deployed in multi-agent systems to overcome context limitations by distributing information across agents. Yet whether agents can reliably co…
Cheap Talk, Empty Promise: Frontier LLMs easily break public promises for self-interest
Jerick Shi, Terry Jingcheng Zhang, Zhijing Jin +1
Large language models are increasingly deployed as autonomous agents in multi-agent settings where they communicate intentions and take consequential actions with limited human ove…
From Sycophancy to Deception: A Unified Taxonomy for LLM Spontaneous Misalignment
Jerick Shi, Terry Jingcheng Zhang, Zhijing Jin +1
Large language models (LLMs) could produce systematically misaligned output, from hallucinated citations to strategic deception of evaluators, yet these phenomena are studied by se…
CLT-Forge: A Scalable Library for Cross-Layer Transcoders and Attribution Graphs
Florent Draye, Vedant Palit, Abir Harrasse +8
Mechanistic interpretability seeks to understand how Large Language Models (LLMs) represent and process information. Recent approaches based on dictionary learning and transcoders…
Fluid Reasoning Representations
Dmitrii Kharlapenko, Terry Jingchen Zhang, Arth Singh +4
Frontier large language models increasingly solve complex tasks involving abstract concepts through extended test-time thinking. Yet we lack a mechanistic account of how extended t…