publications
Full list of publications, sorted by year.
2026
- ICMLThe Deterministic Horizon: When Extended Reasoning Fails and Tool Delegation Becomes NecessaryIn Forty-third International Conference on Machine Learning, ICML 2026, Seoul, South Korea, July 6-11, 2026, 2026To appear
- ICMLPosition: Current XAI Methods Cannot Satisfy Financial AI Explainability RequirementsIn Forty-third International Conference on Machine Learning, ICML 2026, Seoul, South Korea, July 6-11, 2026 - Position Paper Track, 2026To appear
- SIGIRWhen to Retrieve During Reasoning: Adaptive Retrieval for Large Reasoning ModelsIn Proceedings of the 49th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2026, Melbourne, Australia, July 20-24, 2026, 2026To appear
- ACLRouteNLP: Closed-Loop LLM Routing with Conformal Cascading and Distillation Co-OptimizationIn Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 6: Industry Track), ACL 2026, San Diego, CA, USA, July 2-7, 2026, 2026To appear
- ACLAgentEval: DAG-Structured Step-Level Evaluation for Agentic Workflows with Error Propagation TrackingIn Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 6: Industry Track), ACL 2026, San Diego, CA, USA, July 2-7, 2026, 2026To appear
- ACLComplianceNLP: Knowledge-Graph-Augmented RAG for Multi-Framework Regulatory Gap DetectionIn Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 6: Industry Track), ACL 2026, San Diego, CA, USA, July 2-7, 2026, 2026To appear
- ACLFinGround: Detecting and Grounding Financial Hallucinations via Atomic Claim VerificationIn Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 6: Industry Track), ACL 2026, San Diego, CA, USA, July 2-7, 2026, 2026To appear
- HPDCSAGA: Workflow-Atomic Scheduling for AI Agent Inference on GPU ClustersIn Proceedings of the 35th International Symposium on High-Performance Parallel and Distributed Computing, HPDC 2026, Cleveland, OH, USA, July 13-16, 2026, 2026To appear
- ACL CoNLLBrain–LLM Alignment Tracks Training Data, Not TypologyIn Proceedings of the 30th Conference on Computational Natural Language Learning, ACL 2026, San Diego, CA, USA, July 3-4, 2026, 2026To appear
- ACL CoNLLSparse Autoencoders Map Brain–LLM Alignment onto Cortical Semantic TopographyIn Proceedings of the 30th Conference on Computational Natural Language Learning, ACL 2026, San Diego, CA, USA, July 3-4, 2026, 2026To appear
- ACL CoNLLDo Language Models Know What Not to Say? Causal Evidence for Statistical Preemption in LLMsIn Proceedings of the 30th Conference on Computational Natural Language Learning, ACL 2026, San Diego, CA, USA, July 3-4, 2026, 2026To appear
- ACL CoNLLModel Collapse as Cultural EvolutionIn Proceedings of the 30th Conference on Computational Natural Language Learning, ACL 2026, San Diego, CA, USA, July 3-4, 2026, 2026To appear
- GECCOParameter-Efficient Neuroevolution for Diverse LLM Generation: Quality-Diversity Optimization via Prompt Embedding EvolutionIn Genetic and Evolutionary Computation Conference (GECCO), 2026To appear
- GECCOEvoPref: Multi-Objective Evolutionary Optimization Discovers Diverse LLM Alignments Beyond Gradient DescentIn Genetic and Evolutionary Computation Conference (GECCO), 2026To appear
- IEEE CECPredicting When LLMs Enhance Evolutionary Optimization: A Landscape-Aware FrameworkIn IEEE Congress on Evolutionary Computation (CEC), 2026To appear
- IEEE IJCNNOT-Route: Provably Load-Balanced MoE Routing via Optimal TransportIn International Joint Conference on Neural Networks (IJCNN), 2026To appear
- CogSciBias by Necessity: Impossibility Theorems for Sequential Processing with Convergent AI and Human ValidationIn Proceedings of the 48th Annual Meeting of the Cognitive Science Society, CogSci 2026, Rio de Janeiro, Brazil, July 22-25, 2026, 2026To appear
- CogSciThe Interference Gap: Why Human Memory Outperforms RAG Under Semantic CompetitionIn Proceedings of the 48th Annual Meeting of the Cognitive Science Society, CogSci 2026, Rio de Janeiro, Brazil, July 22-25, 2026, 2026To appear
- CogSciWhen Can Human-AI Teams Outperform Individuals? Tight Bounds with Impossibility GuaranteesIn Proceedings of the 48th Annual Meeting of the Cognitive Science Society, CogSci 2026, Rio de Janeiro, Brazil, July 22-25, 2026, 2026To appear
- ICCSProvably Mesh-Independent Neural Operator Preconditioning via Spectral EquivalenceIn Computational Science - ICCS 2026 - 26th International Conference, Hamburg, Germany, June 29 - July 1, 2026, Proceedings, 2026To appear
- ICCSConformalized Neural Operators: Physics-Informed Adaptive Prediction Bands for PDE SurrogatesIn Computational Science - ICCS 2026 - 26th International Conference, Hamburg, Germany, June 29 - July 1, 2026, Proceedings, 2026To appear