Stop Looking for “Important Tokens” in Multimodal Language Models: Duplication Matters More (2025 · 44 citations)
Vision tokens make up most of the input to multimodal LLMs, which makes inference expensive. Most training-free pruning methods score each token's "importance" from attention maps and drop the low scorers. The authors argue this approach has four weaknesses. It treats scores as fixed even after other tokens are…
FinReporting: An Agentic Workflow for Localized Reporting of Cross-Jurisdiction Financial Disclosure (2026 · 3 citations)
Most LLM tools that extract and condense corporate financial statements are built for one market. They assume a single accounting taxonomy and filing format, which fails investors studying foreign firms. Across markets, similarly named line items can differ in meaning, equivalent concepts can carry different labels…
MelTrim: Coarse-to-Fine Data Pruning for Speech Classification (2026)
Data pruning, which means selecting a small coreset that trains models nearly as well as the full dataset, has been studied mostly for images and text.

Co-authors here

12 of 13 co-authors with a page, the most papers together first.

Venues

Where their works here were published, the most works first, with the editions of those that have a page.

Papers of our library cite their works 24 times (of the 149,439 papers whose bibliographies we read); 4 citing papers with a page here →.

Updated 1 November 2026