Adam Fisch

Aarohi Srivastava, Abhinav Rastogi, Abhishek Rao et al. · allen-ai, amazon-science

Language models demonstrate both quantitative improvement and new qualitative capabilities with increasing scale. Despite their potentially transformative impact, these new capabilities are as yet poorly characterized. In order to inform future research, prepare for disruptive new model capabilities, and ameliorate socially harmful effects, it is vital that we understand the present and near-future capabilities and limitations of language models. To address this challenge, we introduce the Beyond the Imitation Game benchmark (BIG-bench). BIG-bench currently consists of 204 tasks, contributed by 450 authors across 132 institutions. Task topics are diverse, drawing problems from linguistics, childhood development, math, common-sense reasoning, biology, physics, social bias, software development, and beyond. BIG-bench focuses on tasks that are believed to be beyond the capabilities of current language models. We evaluate the behavior of OpenAI's GPT models, Google-internal dense transformer architectures, and Switch-style sparse transformers on BIG-bench, across model sizes spanning millions to hundreds of billions of parameters. In addition, a team of human expert raters performed all tasks in order to provide a strong baseline. Findings include: model performance and calibration both improve with scale, but are poor in absolute terms (and when compared with rater performance); performance is remarkably similar across model classes, though with benefits from sparsity; tasks that improve gradually and predictably commonly involve a large knowledge or memorization component, whereas tasks that exhibit "breakthrough" behavior at a critical scale often involve multiple steps or components, or brittle metrics; social bias typically increases with scale in settings with ambiguous context, but this can be improved with prompting.

Victor Quach, Adam Fisch, Tal Schuster et al. · berkeley, mit

We propose a novel approach to conformal prediction for generative language models (LMs). Standard conformal prediction produces prediction sets -- in place of single predictions -- that have rigorous, statistical performance guarantees. LM responses are typically sampled from the model's predicted distribution over the large, combinatorial output space of natural language. Translating this process to conformal prediction, we calibrate a stopping rule for sampling different outputs from the LM that get added to a growing set of candidates until we are confident that the output set is sufficient. Since some samples may be low-quality, we also simultaneously calibrate and apply a rejection rule for removing candidates from the output set to reduce noise. Similar to conformal prediction, we prove that the sampled set returned by our procedure contains at least one acceptable answer with high probability, while still being empirically precise (i.e., small) on average. Furthermore, within this set of candidate responses, we show that we can also accurately identify subsets of individual components -- such as phrases or sentences -- that are each independently correct (e.g., that are not "hallucinations"), again with statistical guarantees. We demonstrate the promise of our approach on multiple tasks in open-domain question answering, text summarization, and radiology report generation using different LM variants.

Tal Schuster, Adam Fisch, Jai Gupta et al. · mit

Recent advances in Transformer-based large language models (LLMs) have led to significant performance improvements across many tasks. These gains come with a drastic increase in the models' size, potentially leading to slow and costly use at inference time. In practice, however, the series of generations made by LLMs is composed of varying levels of difficulty. While certain predictions truly benefit from the models' full capacity, other continuations are more trivial and can be solved with reduced compute. In this work, we introduce Confident Adaptive Language Modeling (CALM), a framework for dynamically allocating different amounts of compute per input and generation timestep. Early exit decoding involves several challenges that we address here, such as: (1) what confidence measure to use; (2) connecting sequence-level constraints to local per-token exit decisions; and (3) attending back to missing hidden representations due to early exits in previous tokens. Through theoretical analysis and empirical experiments on three diverse text generation tasks, we demonstrate the efficacy of our framework in reducing compute -- potential speedup of up to $\times 3$ -- while provably maintaining high performance.

Anastasios N. Angelopoulos, Stephen Bates, Adam Fisch et al. · berkeley, mit

We extend conformal prediction to control the expected value of any monotone loss function. The algorithm generalizes split conformal prediction together with its coverage guarantee. Like conformal prediction, the conformal risk control procedure is tight up to an $\mathcal{O}(1/n)$ factor. We also introduce extensions of the idea to distribution shift, quantile risk control, multiple and adversarial risk control, and expectations of U-statistics. Worked examples from computer vision and natural language processing demonstrate the usage of our algorithm to bound the false negative rate, graph distance, and token-level F1-score.

Adam Fisch, Amal Rannen-Triki, Razvan Pascanu et al. · deepmind

As the application space of language models continues to evolve, a natural question to ask is how we can quickly adapt models to new tasks. We approach this classic question from a continual learning perspective, in which we aim to continue fine-tuning models trained on past tasks on new tasks, with the goal of "transferring" relevant knowledge. However, this strategy also runs the risk of doing more harm than good, i.e., negative transfer. In this paper, we construct a new benchmark of task sequences that target different possible transfer scenarios one might face, such as a sequence of tasks with high potential of positive transfer, high potential for negative transfer, no expected effect, or a mixture of each. An ideal learner should be able to maximally exploit information from all tasks that have any potential for positive transfer, while also avoiding the negative effects of any distracting tasks that may confuse it. We then propose a simple, yet effective, learner that satisfies many of our desiderata simply by leveraging a selective strategy for initializing new models from past task checkpoints. Still, limitations remain, and we hope this benchmark can help the community to further build and analyze such learners.

Adam Fisch, Tommi Jaakkola, Regina Barzilay

Selective classification allows models to abstain from making predictions (e.g., say "I don't know") when in doubt in order to obtain better effective accuracy. While typical selective models can be effective at producing more accurate predictions on average, they may still allow for wrong predictions that have high confidence, or skip correct predictions that have low confidence. Providing calibrated uncertainty estimates alongside predictions -- probabilities that correspond to true frequencies -- can be as important as having predictions that are simply accurate on average. However, uncertainty estimates can be unreliable for certain inputs. In this paper, we develop a new approach to selective classification in which we propose a method for rejecting examples with "uncertain" uncertainties. By doing so, we aim to make predictions with {well-calibrated} uncertainty estimates over the distribution of accepted examples, a property we call selective calibration. We present a framework for learning selectively calibrated models, where a separate selector network is trained to improve the selective calibration error of a given base model. In particular, our work focuses on achieving robust calibration, where the model is intentionally designed to be tested on out-of-domain data. We achieve this through a training strategy inspired by distributionally robust optimization, in which we apply simulated input perturbations to the known, in-domain training data. We demonstrate the empirical effectiveness of our approach on multiple image classification and lung cancer risk assessment tasks.

Bracha Laufer-Goldshtein, Adam Fisch, Regina Barzilay et al.

Machine learning applications frequently come with multiple diverse objectives and constraints that can change over time. Accordingly, trained models can be tuned with sets of hyper-parameters that affect their predictive behavior (e.g., their run-time efficiency versus error rate). As the number of constraints and hyper-parameter dimensions grow, naively selected settings may lead to sub-optimal and/or unreliable results. We develop an efficient method for calibrating models such that their predictions provably satisfy multiple explicit and simultaneous statistical guarantees (e.g., upper-bounded error rates), while also optimizing any number of additional, unconstrained objectives (e.g., total run-time cost). Building on recent results in distribution-free, finite-sample risk control for general losses, we propose Pareto Testing: a two-stage process which combines multi-objective optimization with multiple hypothesis testing. The optimization stage constructs a set of promising combinations on the Pareto frontier. We then apply statistical testing to this frontier only to identify configurations that have (i) high utility with respect to our objectives, and (ii) guaranteed risk levels with respect to our constraints, with specifiable high probability. We demonstrate the effectiveness of our approach to reliably accelerate the execution of large-scale Transformer models in natural language processing (NLP) applications. In particular, we show how Pareto Testing can be used to dynamically configure multiple inter-dependent model attributes -- including the number of layers computed before exiting, number of attention heads pruned, or number of text tokens considered -- to simultaneously control and optimize various accuracy and cost metrics.

Jonathan Berant, Maximillian Chen, Adam Fisch et al.

To handle underspecified or ambiguous queries, AI assistants need a policy for managing their uncertainty to determine (a) when to guess the user intent and answer directly, (b) when to enumerate and answer multiple possible intents, and (c) when to ask a clarifying question. However, such policies are contextually dependent on factors such as user preferences or modality. For example, enumerating multiple possible user intentions is cumbersome on small screens or in a voice setting. In this work, we propose to train steerable policies for managing this uncertainty using self-play. Given two agents, one simulating a user and the other an AI assistant, we generate conversations where the user issues a potentially ambiguous query, and the assistant needs to determine how to respond. Importantly, the model takes as input the numerical cost of each clarification question, and each generated word, and is asked to take the action that will maximize its final reward, which is the cost-penalized accuracy. We use Reinforced Self-Training (ReST) to train our model to achieve high reward and show this leads to a steerable policy that changes its behavior predictably conditioned on the provided costs, leading to higher reward and accuracy. Moreover, our procedure also generalizes to numerical cost values that were unobserved at training time.

Charlie Cowen-Breen, Alekh Agarwal, Stephen Bates et al.

Statistical estimation often involves tradeoffs between expensive, high-quality measurements and a variety of lower-quality proxies. We introduce Multiple-Prediction-Powered Inference (MultiPPI): a general framework for constructing statistically efficient estimates by optimally allocating resources across these diverse data sources. This work provides theoretical guarantees about the minimax optimality, finite-sample performance, and asymptotic normality of the MultiPPI estimator. Through experiments across three diverse large language model (LLM) evaluation scenarios, we show that MultiPPI consistently achieves lower estimation error than existing baselines. This advantage stems from its budget-adaptive allocation strategy, which strategically combines subsets of models by learning their complex cost and correlation structures.

Minghe Shen, Ananth Balashankar, Adam Fisch et al.

The ability to rigorously estimate the failure rates of large language models (LLMs) is a prerequisite for their safe deployment. Currently, however, practitioners often face a tradeoff between expensive human gold standards and potentially severely-biased automatic annotation schemes such as "LLM-as-a-Judge" labeling. In this paper, we propose a new, practical, and efficient approach to LLM failure rate estimation based on constrained maximum-likelihood estimation (MLE). Our method integrates three distinct signal sources: (i) a small, high-quality human-labeled calibration set, (ii) a large corpus of LLM-judge annotations, and, most importantly, (iii) additional side information via domain-specific constraints derived from known bounds on judge performance statistics. We validate our approach through a comprehensive empirical study, benchmarking it against state-of-the-art baselines like Prediction-Powered Inference (PPI). Across diverse experimental regimes -- spanning varying judge accuracies, calibration set sizes, and LLM failure rates -- our constrained MLE consistently delivers more accurate and lower-variance estimates than existing methods. By moving beyond the "black-box" use of automated judges to a flexible framework, we provide a principled, interpretable, and scalable pathway towards LLM failure-rate certification.

Felipe Maia Polo, Aida Nematzadeh, Virginia Aglietti et al.

Moving beyond evaluations that collapse performance across heterogeneous prompts toward fine-grained evaluation at the prompt level, or within relatively homogeneous subsets, is necessary to diagnose generative models' strengths and weaknesses. Such fine-grained evaluations, however, suffer from a data bottleneck: human gold-standard labels are too costly at this scale, while automated ratings are often misaligned with human judgment. To resolve this challenge, we propose a novel statistical model based on tensor factorization that merges cheap autorater data with a limited set of human gold-standard labels. Specifically, our approach uses autorater scores to pretrain latent representations of prompts and generative models, and then aligns those pretrained representations to human preferences using a small calibration set. This sample-efficient methodology is robust to autorater quality, more accurately predicts human preferences on a per-prompt basis than standard baselines, and provides tight confidence intervals for key statistical parameters of interest. We also showcase the practical utility of our method by constructing granular leaderboards based on prompt qualities and by estimating model performance solely from autorater scores, eliminating the need for additional human annotations.

Anthony Liang, Jonathan Berant, Adam Fisch et al.

Reasoning models often spend a significant amount of time thinking before they generate a visible response. In the meantime, they do not give the user any hints as to whether their reasoning is on the right track, and do not give the user any recourse to stop and correct them if their reasoning is flawed. This creates a frustrating, but unfortunately common, experience: the user's time is wasted while the model reasons from a false premise that could have easily been corrected. In contrast, human speakers typically perform lightweight, incremental grounding acts to ensure that participants in the conversation are on the same page; here we ask if language models can learn to leverage a similar type of behavior? With this motivation, we propose interleaved reasoning (IR), in which the model alternates between thinking and surfacing intermediate responses, as an alternative to the standard "think-then-answer" approach. By providing useful information to the user earlier, IR reduces perceived latency, the time a user waits for an initial output, without compromising the quality of the final response. We further introduce a specialization of interleaved reasoning, Plantain (Plan-Thought-Answer Interleaving), where the first intermediate response is an explicit, step-by-step plan for executing the task. This plan-first strategy allows for user intervention and early feedback for subsequent reasoning steps. We demonstrate that Plantain yields an ~6% improvement in pass@1 across several challenging math reasoning and coding benchmarks, while reducing time-to-first-response by over 60% relative to think-then-answer baselines.

Jacob Eisenstein, Chirag Nagpal, Alekh Agarwal et al. · deepmind

Reward models play a key role in aligning language model applications towards human preferences. However, this setup creates an incentive for the language model to exploit errors in the reward model to achieve high estimated reward, a phenomenon often termed \emph{reward hacking}. A natural mitigation is to train an ensemble of reward models, aggregating over model outputs to obtain a more robust reward estimate. We explore the application of reward ensembles to alignment at both training time (through reinforcement learning) and inference time (through reranking). First, we show that reward models are \emph{underspecified}: reward models that perform similarly in-distribution can yield very different rewards when used in alignment, due to distribution shift. Second, underspecification results in overoptimization, where alignment to one reward model does not improve reward as measured by another reward model trained on the same data. Third, overoptimization is mitigated by the use of reward ensembles, and ensembles that vary by their \emph{pretraining} seeds lead to better generalization than ensembles that differ only by their \emph{fine-tuning} seeds, with both outperforming individual reward models. However, even pretrain reward ensembles do not eliminate reward hacking: we show several qualitative reward hacking phenomena that are not mitigated by ensembling because all reward models in the ensemble exhibit similar error patterns.

Namgyu Ho, Sangmin Bae, Taehyeon Kim et al.

We introduce the Block Transformer which adopts hierarchical global-to-local modeling to autoregressive transformers to mitigate the inference bottlenecks associated with self-attention. Self-attention requires the key-value (KV) cache of all previous sequences to be retrieved from memory at every decoding step to retrieve context information, leading to two primary bottlenecks during batch inference. First, there is a significant delay in obtaining the first token, as the information of the entire prompt must first be processed to prefill the KV cache. Second, computation of subsequent tokens is bottlenecked by the high memory I/O demand of fetching the entire KV cache, which grows linearly with sequence length, incurring quadratic memory reads overall. We design the Block Transformer to strategically mitigate these costs, by incorporating coarsity and locality into an integrated global-to-local architecture. At the lower layers, we aggregate tokens into fixed size blocks to apply attention across the entire sequence at coarse-grained detail, to capture the global context while minimizing KV cache overhead. At upper layers, we apply attention within each block to decode individual tokens, to model fine-grained details with a lightweight local KV cache. We pretrain vanilla and Block Transformers from scratch and demonstrate that Block Transformers reach 10--20x inference throughput compared to vanilla transformers with equivalent perplexity and zero-shot task performance. Code is available at https://github.com/itsnamgyu/block-transformer.

Alexander H. Miller, Will Feng, Adam Fisch et al.

We introduce ParlAI (pronounced "par-lay"), an open-source software platform for dialog research implemented in Python, available at http://parl.ai. Its goal is to provide a unified framework for sharing, training and testing of dialog models, integration of Amazon Mechanical Turk for data collection, human evaluation, and online/reinforcement learning; and a repository of machine learning models for comparing with others' models, and improving upon existing architectures. Over 20 tasks are supported in the first release, including popular datasets such as SQuAD, bAbI tasks, MCTest, WikiQA, QACNN, QADailyMail, CBT, bAbI Dialog, Ubuntu, OpenSubtitles and VQA. Several models are integrated, including neural models such as memory networks, seq2seq and attentive LSTMs.

Yasin Abbasi Yadkori, Ilja Kuzborskij, David Stutz et al. · deepmind

We develop a principled procedure for determining when a large language model (LLM) should abstain from responding (e.g., by saying "I don't know") in a general domain, instead of resorting to possibly "hallucinating" a non-sensical or incorrect answer. Building on earlier approaches that use self-consistency as a more reliable measure of model confidence, we propose using the LLM itself to self-evaluate the similarity between each of its sampled responses for a given query. We then further leverage conformal prediction techniques to develop an abstention procedure that benefits from rigorous theoretical guarantees on the hallucination rate (error rate). Experimentally, our resulting conformal abstention method reliably bounds the hallucination rate on various closed-book, open-domain generative question answering datasets, while also maintaining a significantly less conservative abstention rate on a dataset with long responses (Temporal Sequences) compared to baselines using log-probability scores to quantify uncertainty, while achieveing comparable performance on a dataset with short answers (TriviaQA). To evaluate the experiments automatically, one needs to determine if two responses are equivalent given a question. Following standard practice, we use a thresholded similarity function to determine if two responses match, but also provide a method for calibrating the threshold based on conformal prediction, with theoretical guarantees on the accuracy of the match prediction, which might be of independent interest.

Sangmin Bae, Adam Fisch, Hrayr Harutyunyan et al. · mit

Large language models (LLMs) are expensive to deploy. Parameter sharing offers a possible path towards reducing their size and cost, but its effectiveness in modern LLMs remains fairly limited. In this work, we revisit "layer tying" as form of parameter sharing in Transformers, and introduce novel methods for converting existing LLMs into smaller "Recursive Transformers" that share parameters across layers, with minimal loss of performance. Here, our Recursive Transformers are efficiently initialized from standard pretrained Transformers, but only use a single block of unique layers that is then repeated multiple times in a loop. We further improve performance by introducing Relaxed Recursive Transformers that add flexibility to the layer tying constraint via depth-wise low-rank adaptation (LoRA) modules, yet still preserve the compactness of the overall model. We show that our recursive models (e.g., recursive Gemma 1B) outperform both similar-sized vanilla pretrained models (such as TinyLlama 1.1B and Pythia 1B) and knowledge distillation baselines -- and can even recover most of the performance of the original "full-size" model (e.g., Gemma 2B with no shared parameters). Finally, we propose Continuous Depth-wise Batching, a promising new inference paradigm enabled by the Recursive Transformer when paired with early exiting. In a theoretical analysis, we show that this has the potential to lead to significant (2-3x) gains in inference throughput.

Sangmin Bae, Yujin Kim, Reza Bayat et al. · mit

Scaling language models unlocks impressive capabilities, but the accompanying computational and memory demands make both training and deployment expensive. Existing efficiency efforts typically target either parameter sharing or adaptive computation, leaving open the question of how to attain both simultaneously. We introduce Mixture-of-Recursions (MoR), a unified framework that combines the two axes of efficiency inside a single Recursive Transformer. MoR reuses a shared stack of layers across recursion steps to achieve parameter efficiency, while lightweight routers enable adaptive token-level thinking by dynamically assigning different recursion depths to individual tokens. This allows MoR to focus quadratic attention computation only among tokens still active at a given recursion depth, further improving memory access efficiency by selectively caching only their key-value pairs. Beyond these core mechanisms, we also propose a KV sharing variant that reuses KV pairs from the first recursion, specifically designed to further decrease memory footprint. Across model scales ranging from 135M to 1.7B parameters, MoR forms a new Pareto frontier: at equal training FLOPs and smaller model sizes, it significantly lowers validation perplexity and improves few-shot accuracy, while delivering higher throughput compared with vanilla and existing recursive baselines. These gains demonstrate that MoR is an effective path towards large-model quality without incurring large-model cost.

Bracha Laufer-Goldshtein, Adam Fisch, Regina Barzilay et al.

Adjustable hyperparameters of machine learning models typically impact various key trade-offs such as accuracy, fairness, robustness, or inference cost. Our goal in this paper is to find a configuration that adheres to user-specified limits on certain risks while being useful with respect to other conflicting metrics. We solve this by combining Bayesian Optimization (BO) with rigorous risk-controlling procedures, where our core idea is to steer BO towards an efficient testing strategy. Our BO method identifies a set of Pareto optimal configurations residing in a designated region of interest. The resulting candidates are statistically verified and the best-performing configuration is selected with guaranteed risk levels. We demonstrate the effectiveness of our approach on a range of tasks with multiple desiderata, including low error rates, equitable predictions, handling spurious correlations, managing rate and distortion in generative models, and reducing computational costs.

Jacob Eisenstein, Reza Aghajani, Adam Fisch et al.

To be helpful assistants, AI agents must be aware of their own capabilities and limitations. This includes knowing when to answer from parametric knowledge versus using tools, when to trust tool outputs, and when to abstain or hedge. Such capabilities are hard to teach through supervised fine-tuning because they require constructing examples that reflect the agent's specific capabilities. We therefore propose a radically new approach to teaching agents what they know: \emph{collaborative self-play}. We construct multi-agent collaborations in which the group is rewarded for collectively arriving at correct answers. The desired meta-knowledge emerges from the incentives built into the structure of the interaction. We focus on small societies of agents that have access to heterogeneous tools (corpus-specific retrieval), and therefore must collaborate to maximize their success while minimizing their effort. Experiments show that group-level rewards for multi-agent communities can induce policies that \emph{transfer} to improve tool use and selective prediction in settings where individual agents are deployed in isolation.

Dhawal Gupta, Adam Fisch, Christoph Dann et al.

This work tackles the problem of overoptimization in reinforcement learning from human feedback (RLHF), a prevalent technique for aligning models with human preferences. RLHF relies on reward or preference models trained on \emph{fixed preference datasets}, and these models are unreliable when evaluated outside the support of this preference data, leading to the common reward or preference hacking phenomenon. We propose novel, pessimistic objectives for RLHF which are provably robust to overoptimization through the use of pessimism in the face of uncertainty, and design practical algorithms, P3O and PRPO, to optimize these objectives. Our approach is derived for the general preference optimization setting, but can be used with reward models as well. We evaluate P3O and PRPO on the tasks of fine-tuning language models for document summarization and creating helpful assistants, demonstrating remarkable resilience to overoptimization.

Anastasios N. Angelopoulos, Jacob Eisenstein, Jonathan Berant et al. · berkeley

The development lifecycle of generative AI systems requires continual evaluation, data acquisition, and annotation, which is costly in both resources and time. In practice, rapid iteration often makes it necessary to rely on synthetic annotation data because of the low cost, despite the potential for substantial bias. In this paper, we develop novel, cost-aware methods for actively balancing the use of a cheap, but often inaccurate, weak rater -- such as a model-based autorater that is designed to automatically assess the quality of generated content -- with a more expensive, but also more accurate, strong rater alternative such as a human. More specifically, the goal of our approach is to produce a low variance, unbiased estimate of the mean of the target "strong" rating, subject to some total annotation budget. Building on recent work in active and prediction-powered statistical inference, we derive a family of cost-optimal policies for allocating a given annotation budget between weak and strong raters so as to maximize statistical efficiency. Using synthetic and real-world data, we empirically characterize the conditions under which these policies yield improvements over prior methods. We find that, especially in tasks where there is high variability in the difficulty of examples, our policies can achieve the same estimation precision at a far lower total annotation budget than standard evaluation methods.

Adam Fisch, Joshua Maynez, R. Alex Hofer et al.

Prediction-powered inference (PPI) is a method that improves statistical estimates based on limited human-labeled data. PPI achieves this by combining small amounts of human-labeled data with larger amounts of data labeled by a reasonably accurate -- but potentially biased -- automatic system, in a way that results in tighter confidence intervals for certain parameters of interest (e.g., the mean performance of a language model). In this paper, we propose a method called Stratified Prediction-Powered Inference (StratPPI), in which we show that the basic PPI estimates can be considerably improved by employing simple data stratification strategies. Without making any assumptions on the underlying automatic labeling system or data distribution, we derive an algorithm for computing provably valid confidence intervals for population parameters (such as averages) that is based on stratified sampling. In particular, we show both theoretically and empirically that, with appropriate choices of stratification and sample allocation, our approach can provide substantially tighter confidence intervals than unstratified approaches. Specifically, StratPPI is expected to improve in cases where the performance of the autorater varies across different conditional distributions of the target data.

R. Alex Hofer, Joshua Maynez, Bhuwan Dhingra et al.

Prediction-powered inference (PPI) is a method that improves statistical estimates based on limited human-labeled data. Specifically, PPI methods provide tighter confidence intervals by combining small amounts of human-labeled data with larger amounts of data labeled by a reasonably accurate, but potentially biased, automatic system. We propose a framework for PPI based on Bayesian inference that allows researchers to develop new task-appropriate PPI methods easily. Exploiting the ease with which we can design new metrics, we propose improved PPI methods for several importantcases, such as autoraters that give discrete responses (e.g., prompted LLM ``judges'') and autoraters with scores that have a non-linear relationship to human scores.

Adam Fisch, Tal Schuster, Tommi Jaakkola et al.

We develop a new approach to multi-label conformal prediction in which we aim to output a precise set of promising prediction candidates with a bounded number of incorrect answers. Standard conformal prediction provides the ability to adapt to model uncertainty by constructing a calibrated candidate set in place of a single prediction, with guarantees that the set contains the correct answer with high probability. In order to obey this coverage property, however, conformal sets can become inundated with noisy candidates -- which can render them unhelpful in practice. This is particularly relevant to practical applications where there is a limited budget, and the cost (monetary or otherwise) associated with false positives is non-negligible. We propose to trade coverage for a notion of precision by enforcing that the presence of incorrect candidates in the predicted conformal sets (i.e., the total number of false positives) is bounded according to a user-specified tolerance. Subject to this constraint, our algorithm then optimizes for a generalized notion of set coverage (i.e., the true positive rate) that allows for any number of true answers for a given query (including zero). We demonstrate the effectiveness of this approach across a number of classification tasks in natural language processing, computer vision, and computational chemistry.

Tal Schuster, Adam Fisch, Tommi Jaakkola et al.

We develop a novel approach for confidently accelerating inference in the large and expensive multilayer Transformers that are now ubiquitous in natural language processing (NLP). Amortized or approximate computational methods increase efficiency, but can come with unpredictable performance costs. In this work, we present CATs -- Confident Adaptive Transformers -- in which we simultaneously increase computational efficiency, while guaranteeing a specifiable degree of consistency with the original model with high confidence. Our method trains additional prediction heads on top of intermediate layers, and dynamically decides when to stop allocating computational effort to each input using a meta consistency classifier. To calibrate our early prediction stopping rule, we formulate a unique extension of conformal prediction. We demonstrate the effectiveness of this approach on four classification and regression tasks.

Tal Schuster, Adam Fisch, Regina Barzilay

Typical fact verification models use retrieved written evidence to verify claims. Evidence sources, however, often change over time as more information is gathered and revised. In order to adapt, models must be sensitive to subtle differences in supporting evidence. We present VitaminC, a benchmark infused with challenging cases that require fact verification models to discern and adjust to slight factual changes. We collect over 100,000 Wikipedia revisions that modify an underlying fact, and leverage these revisions, together with additional synthetically constructed ones, to create a total of over 400,000 claim-evidence pairs. Unlike previous resources, the examples in VitaminC are contrastive, i.e., they contain evidence pairs that are nearly identical in language and content, with the exception that one supports a given claim while the other does not. We show that training using this design increases robustness -- improving accuracy by 10% on adversarial fact verification and 6% on adversarial natural language inference (NLI). Moreover, the structure of VitaminC leads us to define additional tasks for fact-checking resources: tagging relevant words in the evidence for verifying the claim, identifying factual revisions, and providing automatic edits via factually consistent text generation.

Adam Fisch, Tal Schuster, Tommi Jaakkola et al.

We develop a novel approach to conformal prediction when the target task has limited data available for training. Conformal prediction identifies a small set of promising output candidates in place of a single prediction, with guarantees that the set contains the correct answer with high probability. When training data is limited, however, the predicted set can easily become unusably large. In this work, we obtain substantially tighter prediction sets while maintaining desirable marginal guarantees by casting conformal prediction as a meta-learning paradigm over exchangeable collections of auxiliary tasks. Our conformalization algorithm is simple, fast, and agnostic to the choice of underlying model, learning algorithm, or dataset. We demonstrate the effectiveness of this approach across a number of few-shot classification and regression tasks in natural language processing, computer vision, and computational chemistry for drug discovery.

Tianyu Gao, Adam Fisch, Danqi Chen

The recent GPT-3 model (Brown et al., 2020) achieves remarkable few-shot performance solely by leveraging a natural-language prompt and a few task demonstrations as input context. Inspired by their findings, we study few-shot learning in a more practical scenario, where we use smaller language models for which fine-tuning is computationally efficient. We present LM-BFF--better few-shot fine-tuning of language models--a suite of simple and complementary techniques for fine-tuning language models on a small number of annotated examples. Our approach includes (1) prompt-based fine-tuning together with a novel pipeline for automating prompt generation; and (2) a refined strategy for dynamically and selectively incorporating demonstrations into each context. Finally, we present a systematic evaluation for analyzing few-shot performance on a range of NLP tasks, including classification and regression. Our experiments demonstrate that our methods combine to dramatically outperform standard fine-tuning procedures in this low resource setting, achieving up to 30% absolute improvement, and 11% on average across all tasks. Our approach makes minimal assumptions on task resources and domain expertise, and hence constitutes a strong task-agnostic method for few-shot learning.

Adam Fisch, Kenton Lee, Ming-Wei Chang et al.

The traditional image captioning task uses generic reference captions to provide textual information about images. Different user populations, however, will care about different visual aspects of images. In this paper, we propose a new task, Captioning with a Purpose (CapWAP). Our goal is to develop systems that can be tailored to be useful for the information needs of an intended population, rather than merely provide generic information about an image. In this task, we use question-answer (QA) pairs---a natural expression of information need---from users, instead of reference captions, for both training and post-inference evaluation. We show that it is possible to use reinforcement learning to directly optimize for the intended information need, by rewarding outputs that allow a question answering model to provide correct answers to sampled user questions. We convert several visual question answering datasets into CapWAP datasets, and demonstrate that under a variety of scenarios our purposeful captioning system learns to anticipate and fulfill specific information needs better than its generic counterparts, as measured by QA performance on user questions from unseen images, when using the caption alone as context.

Adam Fisch, Tal Schuster, Tommi Jaakkola et al.

In this paper, we present a novel approach for conformal prediction (CP), in which we aim to identify a set of promising prediction candidates -- in place of a single prediction. This set is guaranteed to contain a correct answer with high probability, and is well-suited for many open-ended classification tasks. In the standard CP paradigm, the predicted set can often be unusably large and also costly to obtain. This is particularly pervasive in settings where the correct answer is not unique, and the number of total possible answers is high. We first expand the CP correctness criterion to allow for additional, inferred "admissible" answers, which can substantially reduce the size of the predicted set while still providing valid performance guarantees. Second, we amortize costs by conformalizing prediction cascades, in which we aggressively prune implausible labels early on by using progressively stronger classifiers -- again, while still providing valid performance guarantees. We demonstrate the empirical effectiveness of our approach for multiple applications in natural language processing and computational chemistry for drug discovery.

Adam Fisch, Alon Talmor, Robin Jia et al.

We present the results of the Machine Reading for Question Answering (MRQA) 2019 shared task on evaluating the generalization capabilities of reading comprehension systems. In this task, we adapted and unified 18 distinct question answering datasets into the same format. Among them, six datasets were made available for training, six datasets were made available for development, and the final six were hidden for final evaluation. Ten teams submitted systems, which explored various ideas including data sampling, multi-task learning, adversarial training and ensembling. The best system achieved an average F1 score of 72.5 on the 12 held-out datasets, 10.7 absolute points higher than our initial baseline based on BERT.

Adam Fisch, Jiang Guo, Regina Barzilay

This paper explores the task of leveraging typology in the context of cross-lingual dependency parsing. While this linguistic information has shown great promise in pre-neural parsing, results for neural architectures have been mixed. The aim of our investigation is to better understand this state-of-the-art. Our main findings are as follows: 1) The benefit of typological information is derived from coarsely grouping languages into syntactically-homogeneous clusters rather than from learning to leverage variations along individual typological dimensions in a compositional manner; 2) Typology consistent with the actual corpus statistics yields better transfer performance; 3) Typological similarity is only a rough proxy of cross-lingual transferability with respect to parsing.

Ledell Wu, Adam Fisch, Sumit Chopra et al.

We present StarSpace, a general-purpose neural embedding model that can solve a wide variety of problems: labeling tasks such as text classification, ranking tasks such as information retrieval/web search, collaborative filtering-based or content-based recommendation, embedding of multi-relational graphs, and learning word, sentence or document level embeddings. In each case the model works by embedding those entities comprised of discrete features and comparing them against each other -- learning similarities dependent on the task. Empirical results on a number of tasks show that StarSpace is highly competitive with existing methods, whilst also being generally applicable to new cases where those methods are not.

Danqi Chen, Adam Fisch, Jason Weston et al.

This paper proposes to tackle open- domain question answering using Wikipedia as the unique knowledge source: the answer to any factoid question is a text span in a Wikipedia article. This task of machine reading at scale combines the challenges of document retrieval (finding the relevant articles) with that of machine comprehension of text (identifying the answer spans from those articles). Our approach combines a search component based on bigram hashing and TF-IDF matching with a multi-layer recurrent neural network model trained to detect answers in Wikipedia paragraphs. Our experiments on multiple existing QA datasets indicate that (1) both modules are highly competitive with respect to existing counterparts and (2) multitask learning using distant supervision on their combination is an effective complete system on this challenging task.

Alexander Miller, Adam Fisch, Jesse Dodge et al.

Directly reading documents and being able to answer questions from them is an unsolved challenge. To avoid its inherent difficulty, question answering (QA) has been directed towards using Knowledge Bases (KBs) instead, which has proven effective. Unfortunately KBs often suffer from being too restrictive, as the schema cannot support certain types of answers, and too sparse, e.g. Wikipedia contains much more information than Freebase. In this work we introduce a new method, Key-Value Memory Networks, that makes reading documents more viable by utilizing different encodings in the addressing and output stages of the memory read operation. To compare using KBs, information extraction or Wikipedia documents directly in a single framework we construct an analysis tool, WikiMovies, a QA dataset that contains raw text alongside a preprocessed KB, in the domain of movies. Our method reduces the gap between all three settings. It also achieves state-of-the-art results on the existing WikiQA benchmark.