Alexis Ross

Zhaofeng Wu, Linlu Qiu, Alexis Ross et al. · microsoft-research, mit

The impressive performance of recent language models across a wide range of tasks suggests that they possess a degree of abstract reasoning skills. Are these skills general and transferable, or specialized to specific tasks seen during pretraining? To disentangle these effects, we propose an evaluation framework based on "counterfactual" task variants that deviate from the default assumptions underlying standard tasks. Across a suite of 11 tasks, we observe nontrivial performance on the counterfactual variants, but nevertheless find that performance substantially and consistently degrades compared to the default conditions. This suggests that while current LMs may possess abstract task-solving skills to an extent, they often also rely on narrow, non-transferable procedures for task-solving. These results motivate a more careful interpretation of language model performance that teases apart these aspects of behavior.

Ian R. McKenzie, Alexander Lyzhov, Michael Pieler et al. · stanford, utoronto

Work on scaling laws has found that large language models (LMs) show predictable improvements to overall loss with increased scale (model size, training data, and compute). Here, we present evidence for the claim that LMs may show inverse scaling, or worse task performance with increased scale, e.g., due to flaws in the training objective and data. We present empirical evidence of inverse scaling on 11 datasets collected by running a public contest, the Inverse Scaling Prize, with a substantial prize pool. Through analysis of the datasets, along with other examples found in the literature, we identify four potential causes of inverse scaling: (i) preference to repeat memorized sequences over following in-context instructions, (ii) imitation of undesirable patterns in the training data, (iii) tasks containing an easy distractor task which LMs could focus on, rather than the harder real task, and (iv) correct but misleading few-shot demonstrations of the task. We release the winning datasets at https://inversescaling.com/data to allow for further investigation of inverse scaling. Our tasks have helped drive the discovery of U-shaped and inverted-U scaling trends, where an initial trend reverses, suggesting that scaling trends are less reliable at predicting the behavior of larger-scale models than previously understood. Overall, our results suggest that there are tasks for which increased model scale alone may not lead to progress, and that more careful thought needs to go into the data and objectives for training language models.

Mike D'Arcy, Alexis Ross, Erin Bransom et al. · allen-ai

We introduce the task of automatically revising scientific papers based on peer feedback and release ARIES, a dataset of review comments and their corresponding paper edits. The data is drawn from real reviewer-author interactions from computer science, and we provide labels linking each reviewer comment to the specific paper edits made by the author in response. We automatically create a high-precision silver training set, as well as an expert-labeled test set that shows high inter-annotator agreement. In experiments with 10 models covering the state of the art, we find that they struggle even to identify which edits correspond to a comment -- especially when the relationship between the edit and the comment is indirect and requires reasoning to uncover. We also extensively analyze GPT-4's ability to generate edits given a comment and the original paper. We find that it often succeeds on a superficial level, but tends to rigidly follow the wording of the feedback rather than the underlying intent, and lacks technical details compared to human-written edits.

Alexis Ross, Matthew E. Peters, Ana Marasović · allen-ai

Rationalization is fundamental to human reasoning and learning. NLP models trained to produce rationales along with predictions, called self-rationalization models, have been investigated for their interpretability and utility to end-users. However, the extent to which training with human-written rationales facilitates learning remains an under-explored question. We ask whether training models to self-rationalize can aid in their learning to solve tasks for the right reasons. Specifically, we evaluate how training self-rationalization models with free-text rationales affects robustness to spurious correlations in fine-tuned encoder-decoder and decoder-only models of six different sizes. We evaluate robustness to spurious correlations by measuring performance on 1) manually annotated challenge datasets and 2) subsets of original test sets where reliance on spurious correlations would fail to produce correct answers. We find that while self-rationalization can improve robustness to spurious correlations in low-resource settings, it tends to hurt robustness in higher-resource settings. Furthermore, these effects depend on model family and size, as well as on rationale content. Together, our results suggest that explainability can come at the cost of robustness; thus, appropriate care should be taken when training self-rationalizing models with the goal of creating more trustworthy models.

Alexis Ross, Megha Srivastava, Jeremiah Blanchard et al.

As programmers write code, they often edit and retry multiple times, creating rich "interaction traces" that reveal how they approach coding tasks and provide clues about their level of skill development. For novice programmers in particular, these traces reflect the diverse reasoning processes they employ to code, such as exploratory behavior to understand how a programming concept works, re-strategizing in response to bugs, and personalizing stylistic choices. In this work, we explore what can be learned from training language models on such reasoning traces: not just about code, but about coders, and particularly students learning to program. We introduce a dataset of over 3.8 million programming reasoning traces from users of Pencil Code, a free online educational platform used by students to learn simple programming concepts. Compared to models trained only on final programs or synthetically-generated traces, we find that models trained on real traces are stronger at modeling diverse student behavior. Through both behavioral and probing analyses, we also find that many properties of code traces, such as goal backtracking or number of comments, can be predicted from learned representations of the students who write them. Building on this result, we show that we can help students recover from mistakes by steering code generation models to identify a sequence of edits that will results in more correct code while remaining close to the original student's style. Together, our results suggest that many properties of code are properties of individual students and that training on edit traces can lead to models that are more steerable, more predictive of student behavior while programming, and better at generating programs in their final states. Code and data is available at https://github.com/meghabyte/pencilcode-public

Belinda Z. Li, Emmy Liu, Alexis Ross et al.

When the world changes, so does the text that humans write about it. How do we build language models that can be easily updated to reflect these changes? One popular approach is retrieval-augmented generation, in which new documents are inserted into a knowledge base and retrieved during prediction for downstream tasks. Most prior work on these systems have focused on improving behavior during prediction through better retrieval or reasoning. This paper introduces ERASE, which instead improves model behavior when new documents are acquired, by incrementally deleting or rewriting other entries in the knowledge base each time a document is added. In two new benchmark datasets evaluating models' ability to answer questions about a stream of news articles or conversations, ERASE improves accuracy relative to conventional retrieval-augmented generation by 7-13% (Mixtral-8x7B) and 6-10% (Llama-3-8B) absolute. Code and data are available at https://github.com/belindal/ERASE

Alexis Ross, Jacob Andreas

When a teacher provides examples for a student to study, these examples must be informative, enabling a student to progress from their current state toward a target concept or skill. Good teachers must therefore simultaneously infer what students already know and adapt their teaching to students' changing state of knowledge. There is increasing interest in using computational models, particularly large language models, as pedagogical tools. As students, language models in particular have shown a remarkable ability to adapt to new tasks given small numbers of examples. But how effectively can these models adapt as teachers to students of different types? To study this question, we introduce a suite of models and evaluation methods we call AdapT. AdapT has two components: (1) a collection of simulated Bayesian student models that can be used for evaluation of automated teaching methods; (2) a platform for evaluation with human students, to characterize the real-world effectiveness of these methods. We additionally introduce (3) AToM, a new probabilistic model for adaptive teaching that jointly infers students' past beliefs and optimizes for the correctness of future beliefs. In evaluations of simulated students across three learning domains (fraction arithmetic, English morphology, function learning), AToM systematically outperforms LLM-based and standard Bayesian teaching models. In human experiments, both AToM and LLMs outperform non-adaptive random example selection. Our results highlight both the difficulty of the adaptive teaching task and the potential of learned adaptive models for solving it.

Shannon Zejiang Shen, Valerie Chen, Ken Gu et al. · cmu

Current evaluations of agents remain centered around one-shot task completion, failing to account for the inherently iterative and collaborative nature of many real-world problems, where human goals are often underspecified and evolve. We argue for a shift from building and assessing task completion agents to developing collaborative agents, assessed not only by the quality of their final outputs but by how well they engage with and enhance human effort throughout the problem-solving process. To support this shift, we introduce collaborative effort scaling, a framework that captures how an agent's utility grows with increasing user involvement. Through case studies and simulated evaluations, we show that state-of-the-art agents often underperform in multi-turn, real-world scenarios, revealing a missing ingredient in agent design: the ability to sustain engagement and scaffold user understanding. Collaborative effort scaling offers a lens for diagnosing agent behavior and guiding development toward more effective interactions.

Alexis Ross, Jacob Andreas

Research on reasoning in language models (LMs) predominantly focuses on improving the correctness of their outputs. But some important applications require modeling reasoning patterns that are incorrect. For example, automated systems that can reason about and simulate student errors are useful for providing real-time feedback in the classroom or offline practice for educators-in-training. This paper presents a new method, MISTAKE, that (1) constructs high-quality synthetic examples of reasoning errors by leveraging cycle consistency between incorrect answers and latent misconceptions; and (2) uses the generated data to learn models for student simulation, misconception classification, and answer generation. We evaluate MISTAKE on three educational tasks and find that it results in (1) higher accuracy when simulating incorrect student answers based on specific misconceptions, (2) increased performance inferring latent misconceptions from observed incorrect answers, and (3) higher alignment with expert-written distractor answers when generating incorrect answers (e.g., for multiple-choice tests).

Canaan Breiss, Alexis Ross, Amani Maina-Kilaas et al. · mit

We propose an interactive approach to language learning that utilizes linguistic acceptability judgments from an informant (a competent language user) to learn a grammar. Given a grammar formalism and a framework for synthesizing data, our model iteratively selects or synthesizes a data-point according to one of a range of information-theoretic policies, asks the informant for a binary judgment, and updates its own parameters in preparation for the next query. We demonstrate the effectiveness of our model in the domain of phonotactics, the rules governing what kinds of sound-sequences are acceptable in a language, and carry out two experiments, one with typologically-natural linguistic data and another with a range of procedurally-generated languages. We find that the information-theoretic policies that our model uses to select items to query the informant achieve sample efficiency comparable to, and sometimes greater than, fully supervised approaches.

Marcos Treviso, Alexis Ross, Nuno M. Guerreiro et al.

Selective rationales and counterfactual examples have emerged as two effective, complementary classes of interpretability methods for analyzing and training NLP models. However, prior work has not explored how these methods can be integrated to combine their complementary advantages. We overcome this limitation by introducing CREST (ContRastive Edits with Sparse raTionalization), a joint framework for selective rationalization and counterfactual text generation, and show that this framework leads to improvements in counterfactual quality, model robustness, and interpretability. First, CREST generates valid counterfactuals that are more natural than those produced by previous methods, and subsequently can be used for data augmentation at scale, reducing the need for human-generated examples. Second, we introduce a new loss function that leverages CREST counterfactuals to regularize selective rationales and show that this regularization improves both model robustness and rationale quality, compared to methods that do not leverage CREST counterfactuals. Our results demonstrate that CREST successfully bridges the gap between selective rationales and counterfactual examples, addressing the limitations of existing methods and providing a more comprehensive view of a model's predictions.

Alexis Ross, Tongshuang Wu, Hao Peng et al.

Controlled text perturbation is useful for evaluating and improving model generalizability. However, current techniques rely on training a model for every target perturbation, which is expensive and hard to generalize. We present Tailor, a semantically-controlled text generation system. Tailor builds on a pretrained seq2seq model and produces textual outputs conditioned on control codes derived from semantic representations. We craft a set of operations to modify the control codes, which in turn steer generation towards targeted attributes. These operations can be further composed into higher-level ones, allowing for flexible perturbation strategies. We demonstrate the effectiveness of these perturbations in multiple applications. First, we use Tailor to automatically create high-quality contrast sets for four distinct natural language processing (NLP) tasks. These contrast sets contain fewer spurious artifacts and are complementary to manually annotated ones in their lexical diversity. Second, we show that Tailor perturbations can improve model generalization through data augmentation. Perturbing just 2% of training data leads to a 5.8-point gain on an NLI challenge set measuring reliance on syntactic heuristics.

Matt Gardner, William Merrill, Jesse Dodge et al.

Much recent work in NLP has documented dataset artifacts, bias, and spurious correlations between input features and output labels. However, how to tell which features have "spurious" instead of legitimate correlations is typically left unspecified. In this work we argue that for complex language understanding tasks, all simple feature correlations are spurious, and we formalize this notion into a class of problems which we call competency problems. For example, the word "amazing" on its own should not give information about a sentiment label independent of the context in which it appears, which could include negation, metaphor, sarcasm, etc. We theoretically analyze the difficulty of creating data for competency problems when human bias is taken into account, showing that realistic datasets will increasingly deviate from competency problems as dataset size increases. This analysis gives us a simple statistical test for dataset artifacts, which we use to show more subtle biases than were described in prior work, including demonstrating that models are inappropriately affected by these less extreme biases. Our theoretical treatment of this problem also allows us to analyze proposed solutions, such as making local edits to dataset instances, and to give recommendations for future data collection and model design efforts that target competency problems.

Alexis Ross, Ana Marasović, Matthew E. Peters

Humans have been shown to give contrastive explanations, which explain why an observed event happened rather than some other counterfactual event (the contrast case). Despite the influential role that contrastivity plays in how humans explain, this property is largely missing from current methods for explaining NLP models. We present Minimal Contrastive Editing (MiCE), a method for producing contrastive explanations of model predictions in the form of edits to inputs that change model outputs to the contrast case. Our experiments across three tasks--binary sentiment classification, topic classification, and multiple-choice question answering--show that MiCE is able to produce edits that are not only contrastive, but also minimal and fluent, consistent with human contrastive edits. We demonstrate how MiCE edits can be used for two use cases in NLP system development--debugging incorrect model outputs and uncovering dataset artifacts--and thereby illustrate that producing contrastive explanations is a promising research direction for model interpretability.

Alexis Ross, Himabindu Lakkaraju, Osbert Bastani

As machine learning models are increasingly deployed in high-stakes domains such as legal and financial decision-making, there has been growing interest in post-hoc methods for generating counterfactual explanations. Such explanations provide individuals adversely impacted by predicted outcomes (e.g., an applicant denied a loan) with recourse -- i.e., a description of how they can change their features to obtain a positive outcome. We propose a novel algorithm that leverages adversarial training and PAC confidence sets to learn models that theoretically guarantee recourse to affected individuals with high probability without sacrificing accuracy. We demonstrate the efficacy of our approach via extensive experiments on real data.

Najoung Kim, Roma Patel, Adam Poliak et al.

We introduce a set of nine challenge tasks that test for the understanding of function words. These tasks are created by structurally mutating sentences from existing datasets to target the comprehension of specific types of function words (e.g., prepositions, wh-words). Using these probing tasks, we explore the effects of various pretraining objectives for sentence encoders (e.g., language modeling, CCG supertagging and natural language inference (NLI)) on the learned representations. Our results show that pretraining on language modeling performs the best on average across our probing tasks, supporting its widespread use for pretraining state-of-the-art NLP models, and CCG supertagging and NLI pretraining perform comparably. Overall, no pretraining objective dominates across the board, and our function word probing tasks highlight several intuitive differences between pretraining objectives, e.g., that NLI helps the comprehension of negation.