Papers by Barun Patra

15 papers
Invariant Language Modeling (2022.emnlp-main)

Copied to clipboard

Challenge: Existing methods to remove spurious correlations and biases involve expensive domain alignment.
Approach: They propose a framework for learning invariant representations that generalize better across environments . they adapt a game-theoretic implementation of IRM to language models .
Outcome: The proposed framework can remove structured noise, ignore correlations and achieve better generalization across environments.
EfficientXLang: Towards Improving Token Efficiency Through Cross-Lingual Reasoning (2025.findings-emnlp)

Copied to clipboard

Challenge: Despite recent advances in Reasoning Language Models, most research focuses solely on English, even though many models are pretrained on multilingual data.
Approach: They evaluate three open-source RLMs: DeepSeek R1, Qwen 2.5, and Qwend 3 across four math datasets and seven typologically diverse languages.
Outcome: The proposed model reduces token usage and preserves accuracy even after translation into English.
To Schedule or not to Schedule: Extracting Task Specific Temporal Entities and Associated Negation Constraints (2020.emnlp-main)

Copied to clipboard

Challenge: Existing models for date-time entity extraction from text are task agnostic, resulting in insufficient results for task specific date-timing extraction.
Approach: They propose a model for extracting subset of date-time entities from text and their negation constraints.
Outcome: The proposed model achieves an absolute gain of 19% f-score points compared to baseline methods in detecting date-time entities relevant to scheduling meetings and a 4% improvement over baseline methods for detecting negation constraints over date-timing entities.
ScopeIt: Scoping Task Relevant Sentences in Documents (2020.coling-industry)

Copied to clipboard

Challenge: a problem faced by conversational agents working with large documents is the frequent presence of information that is irrelevant to the agent.
Approach: They propose a neural model for scoping relevant information from a large document . they show that the model performs better with emails than existing baselines .
Outcome: The proposed model improves intent detection and entity extraction tasks without drop in recall.
Weakly Supervised Attention Networks for Entity Recognition (D19-1)

Copied to clipboard

Challenge: Existing approaches to entity recognition require large amounts of token-level data, which can be expensive and cumbersome to obtain.
Approach: They propose a weakly supervised model that can be annotated at word level from a corpus containing binary presence/absence labels.
Outcome: The proposed model performs reasonably well on the task of entity recognition despite not having access to token-level ground truth data.
A Glitch in the Matrix? Locating and Detecting Language Model Grounding with Fakepedia (2024.acl-long)

Copied to clipboard

Challenge: Large language models (LLMs) have an impressive ability to draw on novel information supplied in their context, yet the mechanisms underlying contextual grounding remain unknown.
Approach: They propose a method to study grounding abilities using a counterfactual dataset constructed to clash with a model's parametric knowledge using Fakepedia.
Outcome: The proposed method evaluates grounding abilities when the internal parametric knowledge clashes with the contextual information.
A Practical Analysis of Human Alignment with *PO (2025.findings-naacl)

Copied to clipboard

Challenge: Prior research focused on identifying the best-performing method to varying hyperparameters . prior research focused primarily on a grid search, which can be impractical for general practitioners .
Approach: They propose a preference optimization method that is more stable across hyperparameters and reduces the average response length.
Outcome: The proposed method increases likelihood of achieving better results through various metrics, such as KL divergence and response length.
Language Model Decoding as Likelihood–Utility Alignment (2023.findings-eacl)

Copied to clipboard

Challenge: Existing studies only compare decoding algorithms in narrow scenarios, and their findings do not generalize across tasks.
Approach: They propose a taxonomy of misalignment mitigation strategies to provide a unifying view of decoding as a tool for alignment.
Outcome: The proposed taxonomy combines likelihood and utility assumptions to provide general statements about decoding as a tool for alignment across tasks.
Scaling Laws for Multilingual Language Models (2025.findings-acl)

Copied to clipboard

Challenge: Existing scaling laws for language models are limited to a limited number of languages, but they can be applied to arbitrary number of different languages.
Approach: They propose a scaling law for general-purpose decoder-only language models trained on multilingual data that shifts focus from individual languages to language families.
Outcome: The proposed scaling law can be applied to models trained on multilingual data . it can be used to predict performance across multiple languages and models .
On the Adaptation of Unlimiformer for Decoder-Only Transformers (2024.lrec-main)

Copied to clipboard

Challenge: despite efforts in the community, most common models have a context length of 4k or less.
Approach: They propose to adapt a vector-retrieval augmentation method to decoder-only transformers . they also expand the experimental setup on summarization to include a new task and an instruction-tuned model .
Outcome: The proposed model performs on par with a model with 2x the context length.
Bilingual Lexicon Induction with Semi-supervision in Non-Isometric Embedding Spaces (P19-1)

Copied to clipboard

Challenge: Recent work on bilingual lexicon induction (BLI) relies on an assumption about the isometry of two embedding spaces.
Approach: They propose a semi-supervised approach that relaxes the isometric assumption while leveraging limited aligned bilingual lexicons and a larger set of unaligned word embeddings.
Outcome: The proposed method obtains state-of-the-art results on 15 of 18 language pairs on the MUSE dataset and does particularly well when the embedding spaces don’t appear isometric.
Beyond English-Centric Bitexts for Better Multilingual Language Representation Learning (2023.acl-long)

Copied to clipboard

Challenge: XY-LENT: X-Y bitext enhanced Language ENcodings achieves state-of-the-art performance over 5 cross-lingual tasks within all model size bands.
Approach: They propose a method for building multilingual representation models that are competitive with existing models and more parameter efficient.
Outcome: The proposed model outperforms XLM-R XXL and is 5x and 6x smaller respectively.
A Length-Extrapolatable Transformer (2023.acl-long)

Copied to clipboard

Challenge: Existing Transformers can only deal with the in-distribution size of inputs.
Approach: They propose a relative position embedding to explicitly maximize attention resolution . they also use blockwise causal attention during inference for better resolution a .
Outcome: The proposed model achieves strong performance in interpolation and extrapolation settings.
Everything you need to know about Multilingual LLMs: Towards fair, performant and reliable models for languages of the world (2023.acl-tutorials)

Copied to clipboard

Challenge: Responsible AI issues such as fairness, bias and toxicity will be discussed in this tutorial .
Approach: This tutorial will describe various aspects of scaling up language technologies to many of the world’s languages by describing the latest research in Massively Multilingual Language Models (MMLMs).
Outcome: This tutorial will cover various aspects of scaling up language technologies to many of the world's languages by describing the latest research in multilingual models.
On Efficiently Acquiring Annotations for Multilingual Models (2022.acl-short)

Copied to clipboard

Challenge: a recent study shows that joint learning across multiple languages performs better than the aforementioned approaches . traditional approaches to support NLP tasks require a lot of annotations to perform . a new approach is to train a model for each language with annotation budget divided equally among them .
Approach: They propose a method for joint learning across multiple languages using a single model . they show that active learning provides additional, complementary benefits .
Outcome: The proposed method outperforms other models on a diverse set of tasks . it can arbitrate its annotation budget to query languages it is less certain on .

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