Search papers, labs, and topics across Lattice.
6
0
10
Offline model-based optimization is fundamentally a ranking problem, and focusing on ranking near-optimal designs beats traditional regression-based surrogate modeling.
Programmers spend more time looking at code tokens that have many semantically similar neighbors, suggesting that semantic context significantly influences attention during code comprehension.
Shrinking visual document retrieval storage by 95% is now possible without sacrificing accuracy, thanks to a layout-aware parsing strategy.
Human eye-tracking data can significantly boost LLM code summarization performance, improving BLEU-4 scores by over 13% via a lightweight attention module that distills gaze patterns into learned priors.
The first comprehensive survey of Visual Document Retrieval reveals how MLLMs are reshaping the field, highlighting the shift towards RAG and agentic systems for complex document understanding.
RLVR's success in long-horizon reasoning hinges on a smooth difficulty spectrum, where mastering easier sub-problems unlocks the ability to tackle harder ones, avoiding frustrating grokking plateaus.
