Optimizing Language Models for Human Preferences is a Causal Inference Problem

As large language models (LLMs) see greater use in academic and commercial settings, there is increasing interest in methods that allow language models to generate texts aligned with human preferences. In this paper, we present an initial exploration of language model optimization for human preferences from direct outcome datasets, where each sample consists of a text and an associated numerical outcome measuring the reader's response. We first propose that language model optimization should be viewed as a causal problem to ensure that the model correctly learns the relationship between the text and the outcome. We formalize this causal language optimization problem, and we develop a method--causal preference optimization (CPO)--that solves an unbiased surrogate objective for the problem. We further extend CPO with doubly robust CPO (DR-CPO), which reduces the variance of the surrogate objective while retaining provably strong guarantees on bias. Finally, we empirically demonstrate the effectiveness of (DR-)CPO in optimizing state-of-the-art LLMs for human preferences on direct outcome data, and we validate the robustness of DR-CPO under difficult confounding conditions.

Text-Transport: Toward Learning Causal Effects of Natural Language

As language technologies gain prominence in real-world settings, it is important to understand how changes to language affect reader perceptions. This can be formalized as the causal effect of varying a linguistic attribute (e.g., sentiment) on a reader's response to the text. In this paper, we introduce Text-Transport, a method for estimation of causal effects from natural language under any text distribution. Current approaches for valid causal effect estimation require strong assumptions about the data, meaning the data from which one can estimate valid causal effects often is not representative of the actual target domain of interest. To address this issue, we leverage the notion of distribution shift to describe an estimator that transports causal effects between domains, bypassing the need for strong assumptions in the target domain. We derive statistical guarantees on the uncertainty of this estimator, and we report empirical results and analyses that support the validity of Text-Transport across data settings. Finally, we use Text-Transport to study a realistic setting--hate speech on social media--in which causal effects do shift significantly between text domains, demonstrating the necessity of transport when conducting causal inference on natural language.

In-Context Pretraining: Language Modeling Beyond Document Boundaries

Large language models (LMs) are currently trained to predict tokens given document prefixes, enabling them to directly perform long-form generation and prompting-style tasks which can be reduced to document completion. Existing pretraining pipelines train LMs by concatenating random sets of short documents to create input contexts but the prior documents provide no signal for predicting the next document. We instead present In-Context Pretraining, a new approach where language models are pretrained on a sequence of related documents, thereby explicitly encouraging them to read and reason across document boundaries. We can do In-Context Pretraining by simply changing the document ordering so that each context contains related documents, and directly applying existing pretraining pipelines. However, this document sorting problem is challenging. There are billions of documents and we would like the sort to maximize contextual similarity for every document without repeating any data. To do this, we introduce approximate algorithms for finding related documents with efficient nearest neighbor search and constructing coherent input contexts with a graph traversal algorithm. Our experiments show In-Context Pretraining offers a simple and scalable approach to significantly enhance LMs'performance: we see notable improvements in tasks that require more complex contextual reasoning, including in-context learning (+8%), reading comprehension (+15%), faithfulness to previous contexts (+16%), long-context reasoning (+5%), and retrieval augmentation (+9%).

SenteCon: Leveraging Lexicons to Learn Human-Interpretable Language Representations

Although deep language representations have become the dominant form of language featurization in recent years, in many settings it is important to understand a model's decision-making process. This necessitates not only an interpretable model but also interpretable features. In particular, language must be featurized in a way that is interpretable while still characterizing the original text well. We present SenteCon, a method for introducing human interpretability in deep language representations. Given a passage of text, SenteCon encodes the text as a layer of interpretable categories in which each dimension corresponds to the relevance of a specific category. Our empirical evaluations indicate that encoding language with SenteCon provides high-level interpretability at little to no cost to predictive performance on downstream tasks. Moreover, we find that SenteCon outperforms existing interpretable language representations with respect to both its downstream performance and its agreement with human characterizations of the text.

Counterfactual Augmentation for Multimodal Learning Under Presentation Bias

In real-world machine learning systems, labels are often derived from user behaviors that the system wishes to encourage. Over time, new models must be trained as new training examples and features become available. However, feedback loops between users and models can bias future user behavior, inducing a presentation bias in the labels that compromises the ability to train new models. In this paper, we propose counterfactual augmentation, a novel causal method for correcting presentation bias using generated counterfactual labels. Our empirical evaluations demonstrate that counterfactual augmentation yields better downstream performance compared to both uncorrected models and existing bias-correction methods. Model analyses further indicate that the generated counterfactuals align closely with true counterfactuals in an oracle setting.

SeedBERT: Recovering Annotator Rating Distributions from an Aggregated Label

Many machine learning tasks -- particularly those in affective computing -- are inherently subjective. When asked to classify facial expressions or to rate an individual's attractiveness, humans may disagree with one another, and no single answer may be objectively correct. However, machine learning datasets commonly have just one "ground truth" label for each sample, so models trained on these labels may not perform well on tasks that are subjective in nature. Though allowing models to learn from the individual annotators' ratings may help, most datasets do not provide annotator-specific labels for each sample. To address this issue, we propose SeedBERT, a method for recovering annotator rating distributions from a single label by inducing pre-trained models to attend to different portions of the input. Our human evaluations indicate that SeedBERT's attention mechanism is consistent with human sources of annotator disagreement. Moreover, in our empirical evaluations using large language models, SeedBERT demonstrates substantial gains in performance on downstream subjective tasks compared both to standard deep learning models and to other current models that account explicitly for annotator disagreement.

Stage-wise Fine-tuning for Graph-to-Text Generation

Graph-to-text generation has benefited from pre-trained language models (PLMs) in achieving better performance than structured graph encoders. However, they fail to fully utilize the structure information of the input graph. In this paper, we aim to further improve the performance of the pre-trained language model by proposing a structured graph-to-text model with a two-step fine-tuning mechanism which first fine-tunes the model on Wikipedia before adapting to the graph-to-text generation. In addition to using the traditional token and position embeddings to encode the knowledge graph (KG), we propose a novel tree-level embedding method to capture the inter-dependency structures of the input graph. This new approach has significantly improved the performance of all text generation metrics for the English WebNLG 2017 dataset.

Context-Dependent Models for Predicting and Characterizing Facial Expressiveness

In recent years, extensive research has emerged in affective computing on topics like automatic emotion recognition and determining the signals that characterize individual emotions. Much less studied, however, is expressiveness, or the extent to which someone shows any feeling or emotion. Expressiveness is related to personality and mental health and plays a crucial role in social interaction. As such, the ability to automatically detect or predict expressiveness can facilitate significant advancements in areas ranging from psychiatric care to artificial social intelligence. Motivated by these potential applications, we present an extension of the BP4D+ dataset with human ratings of expressiveness and develop methods for (1) automatically predicting expressiveness from visual data and (2) defining relationships between interpretable visual signals and expressiveness. In addition, we study the emotional context in which expressiveness occurs and hypothesize that different sets of signals are indicative of expressiveness in different contexts (e.g., in response to surprise or in response to pain). Analysis of our statistical models confirms our hypothesis. Consequently, by looking at expressiveness separately in distinct emotional contexts, our predictive models show significant improvements over baselines and achieve comparable results to human performance in terms of correlation with the ground truth.