ES
Elias Stengel-Eskin
cs.CLcs.AIcs.LGcs.CVcs.CYcs.MAcs.SEastro-ph.GAastro-ph.IMcs.CR
On Valency
published · living versionsW_zn33pdmn·v1 · currentpublished
Vision-Language Model Ensembles Achieve Human-Expert Accuracy for Galaxy Merger Classification
with Marco Chiaberge, Massimo Stiavelli, Colin Norman
1 version
W_8v3chzha·v1published
Preprints & journals
90 papers in the corpus · 2019–2026ToM2Vec: Continuous Theory-of-Mind Tokens for Multi-Agent Coordination2610.29934v1 · Nithin Sivakumaran, Ali Payani, Hyunji Lee et al.2026 · 0 citationsarXivon Valency
ActICL: Post-Training for Agentic Multimodal In-Context Learning2610.19475v1 · Yanshu Li, Ruixiang Tang, Junyi Jessy Li et al.2026 · 0 citationsarXiv
Balancing Memory Pathways: Analyzing and Improving Memory Utilization in Hybrid LMs2610.06750v1 · Hyunji Lee, Joykirat Singh, Zaid Khan et al.2026 · 0 citationsarXiv
TeleTune: Evolving Agent Skills From Offline Telemetry2610.05437v2 · Justin Chih-Yao Chen, Elias Stengel-Eskin, Yan Chen et al.2026 · 0 citationsarXiv
Evaluating Persistent Calibration under Evolving Model Knowledge2609.38797v1 · Victor Wang, Thomas Hofweber, Mohit Bansal et al.2026 · 0 citationsarXiv
FrankenReport: Early Exiting in Long-Form Generation Using Expected Value of Computation2609.06320v1 · Zhengping Jiang, Gonzalo Ramos, Jina Suh et al.2026 · 0 citationsarXiv
GlossoGen: Emergent Language in Complex Multi-Agent LLM Interactions2609.01491v1 · Elias Stengel-Eskin, Newton Sander, Carlos Bonetti et al.2026 · 0 citationsarXiv
Who is the Agent to Blame? Localizing Faithfulness and Citation Mistakes in Agentic Deep Research2608.24306v1 · Eran Hirsch, David Wan, Han Wang et al.2026 · 0 citationsarXiv
CreativeInstruct: Scalably Teaching LLMs to Balance Quality, Creativity, and Diversity2608.07460v1 · Ananya Sahu, Mohit Bansal, Elias Stengel-Eskin2026 · 0 citationsarXiv
Physics Question Scene Graph: Fine-grained Evaluation of Physical Plausibility in Text-to-Video Generation2606.25306v1 · Atin Pothiraj, Jaemin Cho, Yue Zhang et al.2026 · 0 citationsarXiv
CalVerT: Augmenting Agents with Calibrated Verifier Telemetry Improves Action and Learning in Knowledge-Intensive Tasks2606.21777v1 · Ashwin Vinod, Ying Ding, Elias Stengel-Eskin2026 · 0 citationsarXiv
PragReST: Self-Reinforcing Counterfactual Reasoning for Pragmatic Language Understanding2606.18624v2 · Jihyung Park, Minchao Huang, Leqi Liu et al.2026 · 0 citationsarXiv
A History-Aware Visually Grounded Critic for Computer Use Agents2606.11078v1 · Jaewoo Lee, Zaid Khan, Archiki Prasad et al.2026 · 0 citationsarXiv
GPU Forecasters: Language Models as Selective Surrogates for Kernel Runtime Optimization2605.31464v1 · Zaid Khan, Justin Chih-Yao Chen, Jaemin Cho et al.2026 · 0 citationsarXiv
Vision-Language Model Ensembles Achieve Human-Expert Accuracy for Galaxy Merger Classification2606.00415v2 · Marco Chiaberge, Elias Stengel-Eskin, Massimo Stiavelli et al.2026 · 0 citationsarXivon Valency
AVSD: Adaptive-View Self-Distillation by Balancing Consensus and Teacher-Specific Privileged Signals2605.20643v1 · Duy Nguyen, Hanqi Xiao, Archiki Prasad et al.2026 · 0 citationsarXiv
MINTEval: Evaluating Memory under Multi-Target Interference in Long-Horizon Agent Systems2605.18565v2 · Hyunji Lee, Justin Chih-Yao Chen, Joykirat Singh et al.2026 · 0 citationsarXiv
Agent-BRACE: Decoupling Beliefs from Actions in Long-Horizon Tasks via Verbalized State Uncertainty2605.11436v1 · Joykirat Singh, Zaid Khan, Archiki Prasad et al.2026 · 0 citationsarXiv
Diagnosing Capability Gaps in Fine-Tuning Data2604.27547v1 · Saeid Asgari Taghanaki, Rakshanda Agarwal, Bruce Sun et al.2026 · 0 citationsarXiv
MERRIN: A Benchmark for Multimodal Evidence Retrieval and Reasoning in Noisy Web Environments2604.13418v1 · Han Wang, David Wan, Hyunji Lee et al.2026 · 0 citationsarXiv
Playing Along: Learning a Double-Agent Defender for Belief Steering via Theory of Mind2604.11666v2 · Hanqi Xiao, Vaidehi Patil, Zaid Khan et al.2026 · 0 citationsarXiv
Cog-DRIFT: Exploration on Adaptively Reformulated Instances Enables Learning from Hard Reasoning Problems2604.04767v1 · Justin Chih-Yao Chen, Archiki Prasad, Zaid Khan et al.2026 · 0 citationsarXiv
Does AI See like Art Historians? Interpreting How Vision Language Models Recognize Artistic Style2603.11024v3 · Marvin Limpijankit, Milad Alshomary, Yassin Oulad Daoud et al.2026 · 0 citationsarXiv
Evaluating Stochasticity in Deep Research Agents2602.23271v1 · Haotian Zhai, Elias Stengel-Eskin, Pratik Patil et al.2026 · 0 citationsarXiv
Balancing Faithfulness and Performance in Reasoning via Multi-Listener Soft Execution2602.16154v1 · Nithin Sivakumaran, Shoubin Yu, Hyunji Lee et al.2026 · 0 citationsarXiv
Multimodal Fact-Level Attribution for Verifiable Reasoning2602.11509v2 · David Wan, Han Wang, Ziyang Wang et al.2026 · 0 citationsarXiv
Reliable and Responsible Foundation Models: A Comprehensive Survey2602.08145v1 · Xinyu Yang, Junlin Han, Rishi Bommasani et al.2026 · 0 citationsarXiv
Conflict-Resolving and Sharpness-Aware Minimization for Generalized Knowledge Editing with Multiple Updates2602.03696v1 · Duy Nguyen, Hanqi Xiao, Archiki Prasad et al.2026 · 0 citationsarXiv
Routing with Generated Data: Annotation-Free LLM Skill Estimation and Expert Selection2601.09692v1 · Tianyi Niu, Justin Chih-Yao Chen, Genta Indra Winata et al.2026 · 0 citationsarXiv
DART: Leveraging Multi-Agent Disagreement for Tool Recruitment in Multimodal Reasoning2512.07132v1 · Nithin Sivakumaran, Justin Chih-Yao Chen, David Wan et al.2025 · 1 citationarXiv
PRInTS: Reward Modeling for Long-Horizon Information Seeking2511.19314v2 · Jaewoo Lee, Archiki Prasad, Justin Chih-Yao Chen et al.2025 · 0 citationsarXiv
Gistify! Codebase-Level Understanding via Runtime Execution2510.26790v1 · Hyunji Lee, Minseon Kim, Chinmay Singh et al.2025 · 0 citationsarXiv
PoSh: Using Scene Graphs To Guide LLMs-as-a-Judge For Detailed Image Descriptions2510.19060v3 · Amith Ananthram, Elias Stengel-Eskin, Lorena A. Bradford et al.2025 · 0 citationsarXiv
One Life to Learn: Inferring Symbolic World Models for Stochastic Environments from Unguided Exploration2510.12088v2 · Zaid Khan, Archiki Prasad, Elias Stengel-Eskin et al.2025 · 0 citationsarXiv
Think Right: Learning to Mitigate Under-Over Thinking via Adaptive, Attentive Compression2510.01581v2 · Joykirat Singh, Justin Chih-Yao Chen, Archiki Prasad et al.2025 · 0 citationsarXiv
Generalized Correctness Models: Learning Calibrated and Model-Agnostic Correctness Predictors from Historical Patterns2509.24988v2 · Hanqi Xiao, Vaidehi Patil, Hyunji Lee et al.2025 · 0 citationsarXiv
Calibrating Verbalized Confidence with Self-Generated Distractors2509.25532v2 · Victor Wang, Elias Stengel-Eskin2025 · 0 citationsarXiv
The Sum Leaks More Than Its Parts: Compositional Privacy Risks and Mitigations in Multi-Agent Collaboration2509.14284v1 · Vaidehi Patil, Elias Stengel-Eskin, Mohit Bansal2025 · 0 citationsarXiv
Language Models Identify Ambiguities and Exploit Loopholes2508.19546v2 · Jio Choi, Mohit Bansal, Elias Stengel-Eskin2025 · 0 citationsarXiv
RotBench: Evaluating Multimodal Large Language Models on Identifying Image Rotation2508.13968v3 · Tianyi Niu, Jaemin Cho, Elias Stengel-Eskin et al.2025 · 1 citationarXiv
GrAInS: Gradient-based Attribution for Inference-Time Steering of LLMs and VLMs2507.18043v2 · Duy Nguyen, Archiki Prasad, Elias Stengel-Eskin et al.2025 · 1 citationarXiv
Instruction Tuning with and without Context: Behavioral Shifts and Downstream Impact2506.15480v2 · Hyunji Lee, Seunghyun Yoon, Yunjae Won et al.2025 · 0 citationsarXiv
GenerationPrograms: Fine-grained Attribution with Executable Programs2506.14580v1 · David Wan, Eran Hirsch, Elias Stengel-Eskin et al.2025 · 0 citationsarXiv
CLaMR: Contextualized Late-Interaction for Multimodal Content Retrieval2506.06144v1 · David Wan, Han Wang, Elias Stengel-Eskin et al.2025 · 1 citationarXiv
How Well Do Vision-Language Models Understand Movies?2506.06275v2 · Emmanouil Zaranis, Ant'onio Farinhas, Saul Santos et al.2025 · 0 citationsarXiv
CLATTER: Comprehensive Entailment Reasoning for Hallucination Detection2506.05243v1 · Ron Eliav, Arie Cattan, Eran Hirsch et al.2025 · 0 citationsarXiv
LAQuer: Localized Attribution Queries in Content-grounded Generation2506.01187v1 · Eran Hirsch, Aviv Slobodkin, David Wan et al.2025 · 3 citationsarXiv
CAPTURe: Evaluating Spatial Reasoning in Vision Language Models via Occluded Object Counting2504.15485v2 · Atin Pothiraj, Elias Stengel-Eskin, Jaemin Cho et al.2025 · 4 citationsarXiv
Retrieval-Augmented Generation with Conflicting Evidence2504.13079v2 · Han Wang, Archiki Prasad, Elias Stengel-Eskin et al.2025 · 2 citationsarXiv
Executable Functional Abstractions: Inferring Generative Programs for Advanced Math Problems2504.09763v2 · Zaid Khan, Elias Stengel-Eskin, Archiki Prasad et al.2025 · 0 citationsarXiv
Career total: 127 works. 90 are in this corpus.Showing the 50 most recent.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.