Large Language Models have seen expanding application across domains, yet their effectiveness as assistive tools for scientific writing—an endeavor requiring precision, multimodal synthesis, and domain expertise—remains insufficiently understood. We examine the potential of LLMs to support domain experts in scientific writing, with a focus on abstract composition. We design an incentivized randomized controlled trial with a hypothetical conference setup where participants with relevant expertise are split into an author and reviewer pool. Inspired by methods in behavioral science, our novel incentive structure encourages authors to edit the provided abstracts to an acceptable quality for a peer-reviewed submission. Our 2 x 2 between-subject design expands into two dimensions: the implicit source of the provided abstract and the disclosure of it. We find authors make most edits when editing human-written abstracts compared to AI-generated abstracts without source attribution, often guided by higher perceived readability in AI generation. Upon disclosure of source information, the volume of edits converges in both source treatments. Reviewer decisions remain unaffected by the source of the abstract, but bear a significant correlation with the number of edits made. Careful stylistic edits, especially in the case of AI-generated abstracts, in the presence of source information, improve the chance of acceptance. We find that AI-generated abstracts hold potential to reach comparable levels of acceptability to human-written ones with minimal revision, and that perceptions of AI authorship, rather than objective quality, drive much of the observed editing behavior. Our findings reverberate the significance of source disclosure in collaborative scientific writing.

Lester James Validad Miranda, Yizhong Wang, Yanai Elazar, Sachin Kumar, Valentina Pyatkin, Faeze Brahman, Noah A. Smith, Hannaneh Hajishirzi, Pradeep Dasigi. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Patrick Queiroz Da Silva, Hari Sethuraman, Dheeraj Rajagopal, Hannaneh Hajishirzi, Sachin Kumar. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

We introduce TESS 2, a general instructionfollowing diffusion language model that outperforms contemporary instruction-tuned diffusion models, as well as matches and sometimes exceeds strong autoregressive (AR) models.We train TESS 2 by first adapting an AR model via continued pretraining with the usual cross-entropy as diffusion loss, and then performing further instruction tuning.We find that adaptation training as well as the choice of the base model is crucial for training good instruction-following diffusion models.Furthermore, we propose reward guidance, a novel and modular inference-time guidance procedure to align model outputs without needing to train the underlying model.Finally, we show that TESS 2 further improves with increased inference-time compute, highlighting the utility of diffusion LMs in having fine-grained controllability over the amount of compute used at inference time.Code and models are available at https://github.com/hamishivi/tess-2.

Harsh Kohli, Sachin Kumar, Huan Sun. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

Sachin Kumar, Chan Young Park, Yulia Tsvetkov, Noah A. Smith, Hannaneh Hajishirzi. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

Luca Soldaini, Rodney Kinney, Akshita Bhagia, Dustin Schwenk, David Atkinson, Russell Authur, Ben Bogin, Khyathi Chandu, Jennifer Dumas, Yanai Elazar, Valentin Hofmann, Ananya Jha, Sachin Kumar, Li Lucy, Xinxi Lyu, Nathan Lambert, Ian Magnusson, Jacob Morrison, Niklas Muennighoff, Aakanksha Naik, Crystal Nam, Matthew Peters, Abhilasha Ravichander, Kyle Richardson, Zejiang Shen, Emma Strubell, Nishant Subramani, Oyvind Tafjord, Evan Walsh, Luke Zettlemoyer, Noah Smith, Hannaneh Hajishirzi, Iz Beltagy, Dirk Groeneveld, Jesse Dodge, Kyle Lo. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

Language model (LM) prompting—a popular paradigm for solving NLP tasks—has been shown to be susceptible to miscalibration and brittleness to slight prompt variations, caused by its discriminative prompting approach, i.e., predicting the label given the input. To address these issues, we propose Gen-Z—a generative prompting framework for zero-shot text classification. GEN-Z is generative, as it measures the LM likelihood of input text, conditioned on natural language descriptions of labels. The framework is multivariate, as label descriptions allow us to seamlessly integrate additional contextual information about the labels to improve task performance. On various standard classification benchmarks, with six open-source LM families, we show that zero-shot classification with simple contextualization of the data source of the evaluation set consistently outperforms both zero-shot and few-shot baselines while improving robustness to prompt variations. Further, our approach enables personalizing classification in a zero-shot manner by incorporating author, subject, or reader information in the label descriptions.

Xiaochuang Han, Sachin Kumar, Yulia Tsvetkov, Marjan Ghazvininejad. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2024.

Yuhan Liu, Shangbin Feng, Xiaochuang Han, Vidhisha Balachandran, Chan Young Park, Sachin Kumar, Yulia Tsvetkov. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2024.

In multilingual settings, non-Latin scripts and low-resource languages are usually disadvantaged in terms of language models’ utility, efficiency, and cost. Specifically, previous studies have reported multiple modeling biases that the current tokenization algorithms introduce to non-Latin script languages, the main one being over-segmentation. In this work, we propose MAGNET— multilingual adaptive gradient-based tokenization—to reduce over-segmentation via adaptive gradient-based subword tokenization. MAGNET learns to predict segment boundaries between byte tokens in a sequence via sub-modules within the model, which act as internal boundary predictors (tokenizers). Previous gradient-based tokenization methods aimed for uniform compression across sequences by integrating a single boundary predictor during training and optimizing it end-to-end through stochastic reparameterization alongside the next token prediction objective. However, this approach still results in over-segmentation for non-Latin script languages in multilingual settings. In contrast, MAGNET offers a customizable architecture where byte-level sequences are routed through language-script-specific predictors, each optimized for its respective language script. This modularity enforces equitable segmentation granularity across different language scripts compared to previous methods. Through extensive experiments, we demonstrate that in addition to reducing segmentation disparities, MAGNET also enables faster language modeling and improves downstream utility.

Chat-based language models are designed to be helpful, yet they should not comply with every user request. While most existing work primarily focuses on refusal of ``unsafe'' queries, we posit that the scope of noncompliance should be broadened. We introduce a comprehensive taxonomy of contextual noncompliance describing when and how models should not comply with user requests. Our taxonomy spans a wide range of categories including incomplete, unsupported, indeterminate, and humanizing requests (in addition to unsafe requests). To test noncompliance capabilities of language models, we use this taxonomy to develop a new evaluation suite of 1000 noncompliance prompts. We find that most existing models show significantly high compliance rates in certain previously understudied categories with models like GPT-4 incorrectly complying with as many as 30\% of requests.To address these gaps, we explore different training strategies using a synthetically-generated training set of requests and expected noncompliant responses. Our experiments demonstrate that while direct finetuning of instruction-tuned models can lead to both over-refusal and a decline in general capabilities, using parameter efficient methods like low rank adapters helps to strike a good balance between appropriate noncompliance and other capabilities.

We introduce WildTeaming, an automatic red-teaming framework that mines in-the-wild user-chatbot interactions to discover 5.7K unique clusters of novel jailbreak tactics, and then composes selections of multiple mined tactics for systematic exploration of novel and even more challenging jailbreaks. Compared to prior work that performed red-teaming via recruited human workers, gradient-based optimization, or iterative revision with large language models (LLMs), our work investigates jailbreaks from chatbot users in-the-wild who were not specifically instructed to break the system. WildTeaming reveals previously unidentified vulnerabilities of frontier LLMs, resulting in more diverse and successful adversarial attacks compared to state-of-the-art jailbreaking methods. While there exist many datasets for jailbreak evaluation, very few open-source datasets exist for jailbreak training, as safety training data has been closed among all frontier models even when their weights are open. Therefore, with WildTeaming we create WildJailbreak, a large-scale open-source synthetic safety dataset with 262K vanilla (direct request) and adversarial (complex jailbreak) prompt-response pairs. In order to mitigate exaggerated safety behaviors, WildJailbreak provides two contrastive types of queries: 1) harmful queries (both vanilla and adversarial) and 2) benign queries that resemble harmful queries in form but contain no harmful intent. As WildJailbreak considerably upgrades the quality and scale of existing safety resources, it uniquely enables us to examine the scaling effects of data and the interplay of data properties and model capabilities during safety training. Through extensive model training and evaluations, we identify the training properties that enable an ideal balance of safety behaviors: appropriate safeguarding without over-refusal, effective handling of both vanilla and adversarial queries, and minimal, if any, decrease in general capabilities. All the components of WildJailbreak contribute to achieving balanced safety behaviors of models

Despite the growing success of diffusion models in continuous-valued domains (e.g., images), similar efforts for discrete domains such as text have yet to match the performance of autoregressive language models.In this work, we present SSD-LM-a diffusion-based language model with two key design choices.First, SSD-LM is semi-autoregressive, iteratively generating blocks of text, allowing for flexible output length at decoding time while enabling local bidirectional context updates.Second, it is simplex-based, performing diffusion on the natural vocabulary space rather than a learned latent space, allowing us to incorporate classifier guidance and modular control using offthe-shelf classifiers without any adaptation.We evaluate SSD-LM on unconstrained text generation benchmarks, and show that it matches or outperforms strong autoregressive GPT-2 models across standard quality and diversity metrics, while vastly outperforming diffusionbased baselines.On controlled text generation, SSD-LM also outperforms competitive baselines, with an extra advantage in modularity. 1

Tianxing He, Jingyu Zhang, Tianle Wang, Sachin Kumar, Kyunghyun Cho, James Glass, Yulia Tsvetkov. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2023.

Theory of Mind (ToM)-the ability to reason about the mental states of other people-is a key element of our social intelligence.Yet, despite their ever more impressive performance, large-scale neural language models still lack basic theory of mind capabilities out-of-the-box.We posit that simply scaling up models will not imbue them with theory of mind due to the inherently symbolic and implicit nature of the phenomenon, and instead investigate an alternative: can we design a decoding-time algorithm that enhances theory of mind of off-the-shelf neural language models without explicit supervision?We present SYMBOLICTOM, a plug-andplay approach to reason about the belief states of multiple characters in reading comprehension tasks via explicit symbolic representation.More concretely, our approach tracks each entity's beliefs, their estimation of other entities' beliefs, and higher-order levels of reasoning, all through graphical representations, allowing for more precise and interpretable reasoning than previous approaches.Empirical results on the well-known ToMi benchmark (Le et al., 2019) demonstrate that SYMBOLICTOM dramatically enhances off-the-shelf neural networks' theory of mind in a zero-shot setting while showing robust out-of-distribution performance compared to supervised baselines.Our work also reveals spurious patterns in existing theory of mind benchmarks, emphasizing the importance of out-of-distribution evaluation and methods that do not overfit a particular dataset.

Sachin Kumar, Vidhisha Balachandran, Lucille Njoo, Antonios Anastasopoulos, Yulia Tsvetkov. Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics. 2023.

Language models have graduated from being research prototypes to commercialized products offered as web APIs, and recent works have highlighted the multilingual capabilities of these products.The API vendors charge their users based on usage, more specifically on the number of "tokens" processed or generated by the underlying language models.What constitutes a token, however, is training data and model dependent with a large variance in the number of tokens required to convey the same information in different languages.In this work, we analyze the effect of this nonuniformity on the fairness of an API's pricing policy across languages.We conduct a systematic analysis of the cost and utility of OpenAI's language model API on multilingual benchmarks in 22 typologically diverse languages.We show evidence that speakers of a large number of the supported languages are overcharged while obtaining poorer results.These speakers tend to also come from regions where the APIs are less affordable to begin with.Through these analyses, we aim to increase transparency around language model APIs' pricing policies and encourage the vendors to make them more equitable.

Large pretrained language models generate fluent text but are notoriously hard to controllably sample from.In this work, we study constrained sampling from such language models: generating text that satisfies user-defined constraints, while maintaining fluency and model's performance in a downstream task.We propose MUCOLA-a sampling procedure that combines the log-likelihood of the language model with arbitrary (differentiable) constraints in a single energy function, and then generates samples in a non-autoregressive manner.Specifically, it initializes the entire output sequence with noise and follows a Markov chain defined by Langevin Dynamics using the gradients of the energy function.We evaluate MUCOLA on text generation with soft and hard constraints as well as their combinations obtaining significant improvements over competitive baselines for toxicity avoidance, sentiment control, and keyword-guided generation. 1

We present REFEREE, a novel framework for sentence summarization that can be trained reference-free (i.e., requiring no gold summaries for supervision), while allowing direct control for compression ratio.Our work is the first to demonstrate that reference-free, controlled sentence summarization is feasible via the conceptual framework of Symbolic Knowledge Distillation (West et al., 2022), where latent knowledge in pre-trained language models is distilled via explicit examples sampled from the teacher models, further purified with three types of filters: length, fidelity, and Information Bottleneck.Moreover, we uniquely propose iterative distillation of knowledge, where student models from the previous iteration of distillation serve as teacher models in the next iteration.Starting off from a relatively modest set of GPT3-generated summaries, we demonstrate how iterative knowledge distillation can lead to considerably smaller, but better summarizers with sharper controllability.A useful by-product of this iterative distillation process is a high-quality dataset of sentence-summary pairs with varying degrees of compression ratios.Empirical results demonstrate that the final student models vastly outperform the much larger GPT3-Instruct model in terms of the controllability of compression ratios, without compromising the quality of resulting summarization. 1

As large-scale language model pretraining pushes the state-of-the-art in text generation, recent work has turned to controlling attributes of the text such models generate. While modifying the pretrained models via fine-tuning remains the popular approach, it incurs a significant computational cost and can be infeasible due to a lack of appropriate data. As an alternative, we propose \textsc{MuCoCO}---a flexible and modular algorithm for controllable inference from pretrained models. We formulate the decoding process as an optimization problem that allows for multiple attributes we aim to control to be easily incorporated as differentiable constraints. By relaxing this discrete optimization to a continuous one, we make use of Lagrangian multipliers and gradient-descent-based techniques to generate the desired text. We evaluate our approach on controllable machine translation and style transfer with multiple sentence-level attributes and observe significant improvements over baselines.

Attentional, RNN-based encoder-decoder architectures have obtained impressive performance on abstractive summarization of news articles. However, these methods fail to account for long term dependencies within the sentences of a document. This problem is exacerbated in multi-document summarization tasks such as summarizing the popular opinion in threads present in community question answering (CQA) websites such as Yahoo! Answers and Quora. These threads contain answers which often overlap or contradict each other. In this work, we present a hierarchical encoder based on structural attention to model such inter-sentence and inter-document dependencies. We set the popular pointer-generator architecture and some of the architectures derived from it as our baselines and show that they fail to generate good summaries in a multi-document setting. We further illustrate that our proposed model achieves significant improvement over the baseline in both single and multi-document summarization settings -- in the former setting, it beats the baseline by 1.31 and 7.8 ROUGE-1 points on CNN and CQA datasets, respectively; in the latter setting, the performance is further improved by 1.6 ROUGE-1 points on the CQA dataset.

Sachin Kumar, Shuly Wintner, Noah A. Smith, Yulia Tsvetkov. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.

Today electronic communications have become the prime medium for people to express their opinions and influence the policy preferences. One such popular channel reflecting the voice of the masses is electronic petitions. To understand people's perspective on various issues it is important to know what petitions say. However, due to the sheer volume of the petitions it is difficult to process each petition manually. As each petition talks about a different issue, prioritizing on over other is difficult. To alleviate these challenges, we present an end to end system for generating comprehensive and concise summaries from e-petitions. A petition contains multiple aspects, the core problem, evidence(s) in support of the problem and potential solutions. Therefore, it is imperative that an useful summary should contain information about all these three aspects explicitly. To achieve this, our system generates three aspect based summaries for each petition for better understanding. We also introduce a new annotated petition dataset, developed through crowd-sourcing, that served as gold standard. Our model is tested through quantitative and qualitative evaluations.

Automatic short answer grading (ASAG) can reduce tedium for instructors, but is complicated by free-form student inputs. An important ASAG task is to assign ordinal scores to student answers, given some “model” or ideal answers. Here we introduce a novel framework for ASAG by cascading three neural building blocks: Siamese bidirectional LSTMs applied to a model and a student answer, a novel pooling layer based on earth-mover distance (EMD) across all hidden states from both LSTMs, and a flexible final regression layer to output scores. On standard ASAG data sets, our system shows substantial reduction in grade estimation error compared to competitive baselines. We demonstrate that EMD pooling results in substantial accuracy gains, and that a support vector ordinal regression (SVOR) output layer helps outperform softmax. Our system also outperforms recent attention mechanisms on LSTM states.