6 papers
PACT: End-to-End Learning of Human Pose, Contacts, and Forces from Video
Rikhat Akizhanov, Yangsong Zhang, Nikolai Kaliazin +5
Human motion, environmental contacts, and interaction forces are governed by common physical laws, yet existing approaches typically separate visual pose reconstruction from contac…
Closing the Cross-Dialect Gap: Query Plans as a Portable Interface in Text-to-SQL
Corentin Royer, Robin Oester, Yotam Perlitz +3
Text-to-SQL systems are typically trained and evaluated on a single dialect (SQLite), yet production deployments span PostgreSQL, MySQL, ClickHouse, and beyond. We show that this s…
Identify, Locate, Link: End-to-End Key-Value Extraction from Document Images
A. Said Gurbuz, Ahmed Nassar, Christoph Auer +8
Document processing pipelines traditionally cascade optical character recognition (OCR) engines with downstream models for structured information extraction, leading to multi-stage…
DocNavRAG: Document-Structured Graph RAG with Stateful Evidence Construction for Complex Document Question Answering
Dongyang Xie, Yao Tian, Hao Zhang +5
Answering complex questions over large document collections requires assembling complementary evidence across sections and documents. GraphRAG offers structured retrieval but typic…
ThinkBooster: A Unified Framework for Seamless Test-Time Scaling of LLM Reasoning
Vladislav Smirnov, Chieu Nguyen, Sergey Senichev +14
Test-time compute (TTC) scaling has emerged as a powerful paradigm for improving large language model (LLM) reasoning by allocating additional compute during inference, e.g., via m…
FireScope: Wildfire Risk Raster Prediction with a Chain-of-Thought Oracle
Mario Markov, Stefan Maria Ailuro, Luc Van Gool +2
Predicting wildfire risk is a reasoning-intensive spatial problem that requires the integration of visual, climatic, and geographic factors to infer continuous risk maps. Existing…