Publications tagged: #llm

• Feedback Friction: LLMs Struggle to Fully Incorporate External Feedback

  Recent studies have shown LLMs possess some ability to improve their responses when given external feedback. However, it remains unclear how effectively and thoroughly these models can incorporate extrinsic feedback. In an ideal scenario, if LLMs receive near-perfect and complete feedback, we would expect them to fully integrate the feedback and change their incorrect answers to correct ones. In this paper, we systematically investigate LLMs' ability to incorporate feedback by designing a controlled experimental environment. For each problem, a solver model attempts a solution, then a feedback generator with access to near-complete ground-truth answers produces targeted feedback, after which the solver tries again. We evaluate this pipeline across a diverse range of tasks, including math reasoning, knowledge reasoning, scientific reasoning, and general multi-domain evaluations with state-of-the-art language models including Claude 3.7 (with and without extended thinking). Surprisingly, even under these near-ideal conditions, solver models consistently show resistance to feedback, a limitation that we term FEEDBACK FRICTION. To mitigate this limitation, we experiment with sampling-based strategies like progressive temperature increases and explicit rejection of previously attempted incorrect answers, which yield improvements but still fail to help models achieve target performance. We also perform a rigorous exploration of potential causes of FEEDBACK FRICTION, ruling out factors such as model overconfidence and data familiarity. We hope that highlighting this issue in LLMs and ruling out several apparent causes will help future research in self-improvement.

  Dongwei Jiang , Alvin Zhang , Andrew Wang , Nicholas Andrews , Daniel Khashabi

  39th Conference on Neural Information Processing Systems (NeurIPS), 2025

  

  #llm #agents

• Learning Extrapolative Sequence Transformations from Markov Chains

  Most successful applications of deep learning involve similar training and test conditions. However, tasks such as biological sequence design involve searching for sequences that improve desirable properties beyond previously known values, which requires novel hypotheses that \emph{extrapolate} beyond training data. In these settings, extrapolation may be achieved by using random search methods such as Markov chain Monte Carlo (MCMC), which, given an initial state, sample local transformations to approximate a target density that rewards states with the desired properties. However, even with a well-designed proposal, MCMC may struggle to explore large structured state spaces efficiently. Rather than relying on stochastic search, it would be desirable to have a model that greedily optimizes the properties of interest, successfully extrapolating in as few steps as possible. We propose to learn such a model from the Markov chains resulting from MCMC search. Specifically, our approach uses selected states from Markov chains as a source of training data for an autoregressive model, which is then able to efficiently generate novel sequences that extrapolate along the sequence-level properties of interest. The proposed approach is validated on three problems: protein sequence design, text sentiment control, and text anonymization. We find that the autoregressive model can extrapolate as well or better than MCMC, but with the additional benefits of scalability and significantly higher sample efficiency.

  Sophia Hager , Aleem Khan , Andrew Wang , Nicholas Andrews

  Forty-Second International Conference on Machine Learning (ICML), 2025

  

  #ml #ai4science #llm

• Are Paraphrases Generated by Large Language Models Invertible?

  High-quality paraphrases are easy to produce using instruction-tuned language models or specialized paraphrasing models. Although this capability has a variety of benign applications, paraphrasing attacks—paraphrases applied to machine-generated texts—are known to significantly degrade the performance of machine-text detectors. This motivates us to consider the novel problem of paraphrase inversion, where, given paraphrased text, the objective is to recover an approximation of the original text. The closer the approximation is to the original text, the better machine-text detectors will perform. We propose an approach which frames the problem as translation from paraphrased text back to the original text, which requires examples of texts and corresponding paraphrases to train the inversion model. Fortunately, such training data can easily be generated, given a corpus of original texts and one or more paraphrasing models. We find that language models such as GPT-4 and Llama-3 exhibit biases when paraphrasing which an inversion model can learn with a modest amount of data. Perhaps surprisingly, we also find that such models generalize well, including to paraphrase models unseen at training time. Finally, we show that when combined with a paraphrased-text detector, our inversion models provide an effective defense against paraphrasing attacks, and overall our approach yields an average improvement of +22% AUROC across seven machine-text detectors and three different domains.

  Rafael Rivera-Soto , Barry Chen , Nicholas Andrews

  Findings of the ACL, 2025

  

  #llm #deepfake_detection

• Uncertainty Distillation: Teaching Language Models to Express Semantic Confidence

  As large language models (LLMs) are increasingly used for factual question-answering, it becomes more important for LLMs to have the capability to communicate the likelihood that their answer is correct. For these verbalized expressions of uncertainty to be meaningful, they should reflect the error rates at the expressed level of confidence. However, when prompted to express confidence, the error rates of current LLMs are inconsistent with their communicated confidences, highlighting the need for uncertainty quantification methods. Many prior methods calculate lexical uncertainty, estimating a model's confidence in the specific string it generated. In some cases, however, it may be more useful to estimate semantic uncertainty, or the model's confidence in the answer regardless of how it is verbalized. We propose a simple procedure, uncertainty distillation, to teach an LLM to verbalize calibrated semantic confidences. Using held-out data to map initial uncertainty estimates to meaningful probabilities, we create examples annotated with verbalized probabilities for supervised fine-tuning. We compare uncertainty distillation to several strong baselines, and find that our method yields verbalized confidences that correlate well with observed error rates.

  Sophia Hager , David Mueller , Kevin Duh , Nicholas Andrews

  arXiv preprint arXiv:2503.14749, 2025

  

  #llm #uncertainty #preprint

• Language Models Optimized to Fool Detectors Still Have a Distinct Style (And How to Change It)

  Despite considerable progress in the development of machine-text detectors, it has been suggested that the problem is inherently hard, and therefore, that stakeholders should proceed under the assumption that machine-generated text cannot be reliably detected as such. We examine a recent such claim by Nicks et al. (2024) regarding the ease with which language models can be optimized to degrade the performance of machine-text detectors, including detectors not specifically optimized against. We identify a feature space–the stylistic feature space–that is robust to such optimization, and show that it may be used to reliably detect samples from language models optimized to prevent detection. Furthermore, we show that even when models are explicitly optimized against stylistic detectors, detection performance remains surprisingly unaffected. We then seek to understand if stylistic detectors are inherently more robust. To study this question, we explore a new paraphrasing approach that simultaneously aims to close the gap between human writing and machine writing in stylistic feature space while avoiding detection using traditional features. We show that when only a single sample is available for detection, this attack is universally effective across all detectors considered, including those that use writing style. However, as the number of samples available for detection grows, the human and machine distributions become distinguishable. This observation encourages us to introduce AURA, a metric that estimates the overlap between human and machine-generated distributions by analyzing how detector performance improves as more samples become available. Overall, our findings underscore previous recommendations to avoid reliance on machine-text detection.

  Rafael Rivera Soto , Barry Chen , Nicholas Andrews

  arXiv preprint arXiv:2505.14608, 2025

  

  #llm #deepfake_detection #preprint

• AnaloBench: Benchmarking the Identification of Abstract and Long-context Analogies

  Humans regularly engage in analogical thinking, relating personal experiences to current situations (X is analogous to Y because of Z). Analogical thinking allows humans to solve problems in creative ways, grasp difficult concepts, and articulate ideas more effectively. Can language models (LMs) do the same? To answer this question, we propose AnaloBench, a benchmark to determine analogical reasoning ability in LMs. Our benchmarking approach focuses on aspects of this ability that are common among humans: (i) recalling related experiences from a large amount of information, and (ii) applying analogical reasoning to complex and lengthy scenarios. We collect a set of 340 high quality, human written analogies for use in our benchmark, which constitutes the largest such collection to date. We then test a broad collection of models consisting of 12 open source and 3 proprietary in various sizes and architectures. As in prior results, scaling up LMs results in some performance boosts. Surprisingly, scale offers minimal gains when, (i) analogies involve lengthy scenarios, or (ii) recalling relevant scenarios from a large pool of information, a process analogous to finding a needle in a haystack. We hope these observations encourage further research in this field.

  Xiao Ye , Andrew Wang , Jacob Choi , Yining Lu , Shreya Sharma , Lingfeng Shen , Vijay Murari Tiyyala , Nicholas Andrews , Daniel Khashabi

  Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2024

  

  #benchmark #llm

• Multi-Task Transfer Matters During Instruction-Tuning

  Instruction-tuning trains a language model on hundreds of tasks jointly to improve a model’s ability to learn in-context;however, the mechanisms that drive in-context learning are poorly understood and, as a result, the role of instruction-tuning on in-context generalization is poorly understood as well.In this work, we study the impact of instruction-tuning on multi-task transfer: how well a model’s parameters adapt to an unseen task via fine-tuning.We find that instruction-tuning negatively impacts a model’s transfer to unseen tasks, and that model transfer and in-context generalization are highly correlated, suggesting that this catastrophic forgetting may impact in-context learning.We study methods to improve model transfer, finding that multi-task training—how well the training tasks are optimized—can significantly impact ICL generalization; additionally, we find that continual training on unsupervised pre-training data can mitigate forgetting and improve ICL generalization as well.Finally, we demonstrate that, early into training, the impact of instruction-tuning on model transfer to tasks impacts in-context generalization on that task.Overall, we provide significant evidence that multi-task transfer is deeply connected to a model’s ability to learn a task in-context.

  David Mueller , Mark Dredze , Nicholas Andrews

  Findings of the Association for Computational Linguistics ACL 2024, 2024

  

  #llm

• Few-Shot Detection of Machine-Generated Text using Style Representations

  The advent of instruction-tuned language models that convincingly mimic human writing poses a significant risk of abuse. However, such abuse may be counteracted with the ability to detect whether a piece of text was composed by a language model rather than a human author. Some previous approaches to this problem have relied on supervised methods by training on corpora of confirmed human- and machine- written documents. Unfortunately, model under-specification poses an unavoidable challenge for neural network-based detectors, making them brittle in the face of data shifts, such as the release of newer language models producing still more fluent text than the models used to train the detectors. Other approaches require access to the models that may have generated a document in question, which is often impractical. In light of these challenges, we pursue a fundamentally different approach not relying on samples from language models of concern at training time. Instead, we propose to leverage representations of writing style estimated from human-authored text. Indeed, we find that features effective at distinguishing among human authors are also effective at distinguishing human from machine authors, including state-of-the-art large language models like Llama-2, ChatGPT, and GPT-4. Furthermore, given a handful of examples composed by each of several specific language models of interest, our approach affords the ability to predict which model generated a given document. The code and data to reproduce our experiments are available at this https URL.

  Rafael Rivera-Soto , Kailin Koch , Aleem Khan , Barry Chen , Marcus Bishop , Nicholas Andrews

  International Conference on Learning Representations (ICLR), 2024

  

  #deepfake_detection #llm #ml

• Learning to Generate Text in Arbitrary Writing Styles

  Prior work in style-controlled text generation has focused on tasks such as emulating the style of prolific literary authors, producing formal or informal text, and mitigating toxicity of generated text. Plentiful demonstrations of these styles are available, and as a result modern language models are often able to emulate them, either via prompting or discriminative control. However, in applications such as writing assistants, it is desirable for language models to produce text in an author-specific style on the basis of a potentially small writing sample. For example, someone writing in a particular dialect may prefer writing suggestions that retain the same dialect. We find that instruction-tuned language models can struggle to reproduce author-specific style demonstrated in a prompt. Instead, we propose to guide a language model to generate text in a target style using contrastively-trained representations that capture stylometric features. Our approach (StyleMC) combines an author-adapted language model with sequence-level inference to improve stylistic consistency, and is found to be effective in a variety of conditions, including unconditional generation and style transfer. Additionally, we find that the proposed approach can serve as an effective anonymization method, by editing a document to mask authorship while preserving the original meaning.

  Aleem Khan , Andrew Wang , Sophia Hager , Nicholas Andrews

  arXiv preprint arXiv:2312.17242, 2023

  

  #controllable_generation #llm #preprint

• Do Text-to-Text Multi-Task Learners Suffer from Task Conflict?

  Traditional multi-task learning architectures learn a single model across multiple tasks through a shared encoder followed by task-specific decoders. Learning these models often requires specialized training algorithms that address task-conflict in the shared parameter updates, which otherwise can lead to negative transfer. A new type of multi-task learning within NLP homogenizes multi-task architectures as a shared encoder and language model decoder, which does surprisingly well across a range of diverse tasks. Does this new architecture suffer from task-conflicts that require specialized training algorithms? We study how certain factors in the shift towards text-to-text models affects multi-task conflict and negative transfer, finding that both directional conflict and transfer are surprisingly constant across architectures.

  David Mueller , Nicholas Andrews , Mark Dredze

  Findings of the 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2022

  

  #ml #llm

• Low-Resource Authorship Style Transfer: Can Non-Famous Authors Be Imitated?

  Authorship style transfer involves altering text to match the style of a target author whilst preserving the original meaning. Existing unsupervised approaches like STRAP have largely focused on style transfer to target authors with many examples of their writing style in books, speeches, or other published works. This high-resource training data requirement (often greater than 100,000 words) makes these approaches primarily useful for style transfer to published authors, politicians, or other well-known figures and authorship styles, while style transfer to non-famous authors has not been well-studied. We introduce the low-resource authorship style transfer task, a more challenging class of authorship style transfer where only a limited amount of text in the target author's style may exist. In our experiments, we specifically choose source and target authors from Reddit and style transfer their Reddit posts, limiting ourselves to just 16 posts (on average ~500 words) of the target author's style. Style transfer accuracy is typically measured by how often a classifier or human judge will classify an output as written by the target author. Recent authorship representations models excel at authorship identification even with just a few writing samples, making automatic evaluation of this task possible for the first time through evaluation metrics we propose. Our results establish an in-context learning technique we develop as the strongest baseline, though we find current approaches do not yet achieve mastery of this challenging task. We release our data and implementations to encourage further investigation.

  Ajay Patel , Nicholas Andrews , Chris Callison-Burch

  arXiv preprint arXiv:2212.08986, 2022

  

  #llm #controllable_generation

Back to all publications