Papers by Pramit Chaudhuri

4 papers
KRISTEVA: Close Reading as a Novel Task for Benchmarking Interpretive Reasoning (2025.acl-long)

Copied to clipboard

Challenge: a study of close reading skills in large language models (LLMs) shows that LLMs still lag behind human evaluators on 10 of 11 tasks.
Approach: They propose a benchmark to evaluate close reading skills in large language models . they propose three tasks to approximate different elements of the close reading process .
Outcome: The proposed benchmarks show that state-of-the-art LLMs possess some college-level close reading competency, but performance still trails human evaluators on 10 out of 11 tasks.
AcrosticSleuth: Probabilistic Identification and Ranking of Acrostics in Multilingual Corpora (2025.findings-naacl)

Copied to clipboard

Challenge: acrostics are hidden messages in which initial letters of consecutive lines or paragraphs form meaningful words or phrases.
Approach: They propose a method to identify acrostics automatically and rank them by the probability that the corresponding sequence of characters does not occur by chance.
Outcome: The proposed method achieves F1 scores of 0.39, 0.59, and 0.66 on the French, English, and Russian subdomains of WikiSource.
Profiling of Intertextuality in Latin Literature Using Word Embeddings (2021.naacl-main)

Copied to clipboard

Challenge: a new study examines the intertextual relationships between authors in classical Latin literature . a large corpus of lemmatized Latin is used to train word embeddings .
Approach: They propose to train an optimized word2vec model on a large corpus of Latin . they then replicate a previous study of the Roman historian Livy using hand-crafted stylometric features.
Outcome: The proposed model outperforms a widely used lexical search method on Latin epic poetry . it advances the development of core computational resources for a major premodern language .
A Stylometry Toolkit for Latin Literature (D19-3)

Copied to clipboard

Challenge: a stylometric toolkit for analysis of Latin literary texts is available for free at www.qcrit.org/stylometry.
Approach: They propose a stylometric toolkit for analysis of Latin literary texts which generates data for a diverse range of literary features and has an intuitive point-and-click interface.
Outcome: The proposed toolkit generates data for a diverse range of literary features and has an intuitive point-and-click interface.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations