Structural cues that organize facts by event improve LLM topic clustering and temporal QA on a new streaming-news benchmark, but temporal reasoning and summarization integration remain hard.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Can Structural Cues Save LLMs? Evaluating Language Models in Massive Document Streams
Structural cues that organize facts by event improve LLM topic clustering and temporal QA on a new streaming-news benchmark, but temporal reasoning and summarization integration remain hard.