| 1-G | Filtered Reasoning Score: Evaluating Reasoning Quality on a Model´s Most-Confident Traces | Manas Pathak | |
| 2-I | Guard Vector: Beyond English LLM Guardrails with Task-Vector Composition and Streaming-Aware Prefix SFT | Wonhyuk Lee | |
| 2-G | Monocultural Biases: Correlated biases in large language models lead to unequal systemic exclusion rates in hiring | Matthew Bone | |
| 4-F | How Can We Synthesize High-Quality Pretraining Data? A Systematic Study of Prompt Design, Generator Model, and Source Data | Joel Niklaus | |
| 4-I | Agent Bazaar: Enabling Economic Alignment in Multi-Agent Marketplaces | Seth Karten | |
| 4-F | ScienceMeter: Tracking Scientific Knowledge Updates in Language Models | Yike Wang | |
| 6-G | MetaState: Persistent Working Memory Enhances Reasoning in Discrete Diffusion Language Models | Kejing Xia | |
October 6-9, 2026 San Francisco

Digital Object Identifier. Official code used to identify documents published on internet; similar to ISBN for books. You may use this code to reference your poster in future scientific publications or CVs. It can be found from anywhere in the world. To find the poster page, log onto www.medra.org and enter the DOI, or enter in your internet browser https://dx.doi.org/ followed by the DOI string asigned to your congress. |
||||
|
|