The Flow has not summarised this story yet — read it at arXiv Computation and Language.
arXiv:2609.01432v1 Announce Type: cross Abstract: Scientific citations carry rhetorical intent. Scholars may cite prior work positively (supporting), negatively (contrasting), or neutrally (mentionin...
arXiv:2606. 24894v2 Announce Type: replace-cross Abstract: Large language models have shown strong fluency in scientific writing, yet the evaluation of related work generation (RWG) remains limited.
arXiv:2609.14770v1 Announce Type: cross Abstract: Generalisations are common in scientific communication, even though they are semantically ambiguous. An automated method is needed to identify and ca...
The paper introduces IDRBench, a framework designed to evaluate how well large language models (LLMs) can integrate knowledge across disciplines for interdisciplinary research. It comprises datasets and tasks—IDR Paper Identification, IDR Idea Integration, and IDR Idea Recommendation—to benchmark LLM performance. The authors analyze ten mainstream LLMs, offering a comprehensive assessment and establishing baselines for future studies.
arXiv:2606.22342v2 Announce Type: replace Abstract: How does research evolve, and can we trace it at the level of individual claims? Scientific progress is not simply a uniform accumulation of facts....
How does research evolve, and what substrate would let us forecast where it goes next? Scientific progress is not simply a uniform accumulation of facts: ideas extend prior methods, address known limitations, realize proposed future directions, and sometimes dispute earlier claims.