Markus Dreyer

David Wan, Mengwen Liu, Kathleen McKeown et al. · amazon-science

Despite significant progress in understanding and improving faithfulness in abstractive summarization, the question of how decoding strategies affect faithfulness is less studied. We present a systematic study of the effect of generation techniques such as beam search and nucleus sampling on faithfulness in abstractive summarization. We find a consistent trend where beam search with large beam sizes produces the most faithful summaries while nucleus sampling generates the least faithful ones. We propose two faithfulness-aware generation methods to further improve faithfulness over current generation techniques: (1) ranking candidates generated by beam search using automatic faithfulness metrics and (2) incorporating lookahead heuristics that produce a faithfulness score on the future summary. We show that both generation methods significantly improve faithfulness across two datasets as evaluated by four automatic faithfulness metrics and human evaluation. To reduce computational cost, we demonstrate a simple distillation approach that allows the model to generate faithful summaries with just greedy decoding. Our code is publicly available at https://github.com/amazon-science/faithful-summarization-generation

Amazon AGI, Aaron Langford, Aayush Shah et al. · amazon-science

We present Amazon Nova, a new generation of state-of-the-art foundation models that deliver frontier intelligence and industry-leading price performance. Amazon Nova Pro is a highly-capable multimodal model with the best combination of accuracy, speed, and cost for a wide range of tasks. Amazon Nova Lite is a low-cost multimodal model that is lightning fast for processing images, video, documents and text. Amazon Nova Micro is a text-only model that delivers our lowest-latency responses at very low cost. Amazon Nova Canvas is an image generation model that creates professional grade images with rich customization controls. Amazon Nova Reel is a video generation model offering high-quality outputs, customization, and motion control. Our models were built responsibly and with a commitment to customer trust, security, and reliability. We report benchmarking results for core capabilities, agentic performance, long context, functional adaptation, runtime performance, and human evaluation.

Arthur Bražinskas, Ramesh Nallapati, Mohit Bansal et al. · amazon-science

Abstractive summarization models are typically pre-trained on large amounts of generic texts, then fine-tuned on tens or hundreds of thousands of annotated samples. However, in opinion summarization, large annotated datasets of reviews paired with reference summaries are not available and would be expensive to create. This calls for fine-tuning methods robust to overfitting on small datasets. In addition, generically pre-trained models are often not accustomed to the specifics of customer reviews and, after fine-tuning, yield summaries with disfluencies and semantic mistakes. To address these problems, we utilize an efficient few-shot method based on adapters which, as we show, can easily store in-domain knowledge. Instead of fine-tuning the entire model, we add adapters and pre-train them in a task-specific way on a large corpus of unannotated customer reviews, using held-out reviews as pseudo summaries. Then, fine-tune the adapters on the small available human-annotated dataset. We show that this self-supervised adapter pre-training improves summary quality over standard fine-tuning by 2.0 and 1.3 ROUGE-L points on the Amazon and Yelp datasets, respectively. Finally, for summary personalization, we condition on aspect keyword queries, automatically created from generic datasets. In the same vein, we pre-train the adapters in a query-based manner on customer reviews and then fine-tune them on annotated datasets. This results in better-organized summary content reflected in improved coherence and fewer redundancies.

Jonathan Pilault, Can Liu, Mohit Bansal et al. · amazon-science, mila

Prompts have been shown to be an effective method to adapt a frozen Pretrained Language Model (PLM) to perform well on downstream tasks. Prompts can be represented by a human-engineered word sequence or by a learned continuous embedding. In this work, we investigate conditional and compositional differentiable prompting. We propose a new model, Prompt Production System (PRopS), which learns to transform task instructions or input metadata, into continuous prompts that elicit task-specific outputs from the PLM. Our model uses a modular network structure based on our neural formulation of Production Systems, which allows the model to learn discrete rules -- neural functions that learn to specialize in transforming particular prompt input patterns, making it suitable for compositional transfer learning and few-shot learning. We present extensive empirical and theoretical analysis and show that PRopS consistently surpasses other PLM adaptation techniques, and often improves upon fully fine-tuned models, on compositional generalization tasks, controllable summarization and multilingual translation, while needing fewer trainable parameters.

Leonardo F. R. Ribeiro, Mohit Bansal, Markus Dreyer · amazon-science

Readability refers to how easily a reader can understand a written text. Several factors affect the readability level, such as the complexity of the text, its subject matter, and the reader's background knowledge. Generating summaries based on different readability levels is critical for enabling knowledge consumption by diverse audiences. However, current text generation approaches lack refined control, resulting in texts that are not customized to readers' proficiency levels. In this work, we bridge this gap and study techniques to generate summaries at specified readability levels. Unlike previous methods that focus on a specific readability level (e.g., lay summarization), we generate summaries with fine-grained control over their readability. We develop three text generation techniques for controlling readability: (1) instruction-based readability control, (2) reinforcement learning to minimize the gap between requested and observed readability and (3) a decoding approach that uses lookahead to estimate the readability of upcoming decoding steps. We show that our generation methods significantly improve readability control on news summarization (CNN/DM dataset), as measured by various readability metrics and human judgement, establishing strong baselines for controllable readability in summarization.

Adithya Pratapa, Kevin Small, Markus Dreyer · amazon-science, cmu

Generating concise summaries of news events is a challenging natural language processing task. While journalists often curate timelines to highlight key sub-events, newcomers to a news event face challenges in catching up on its historical context. In this paper, we address this need by introducing the task of background news summarization, which complements each timeline update with a background summary of relevant preceding events. We construct a dataset by merging existing timeline datasets and asking human annotators to write a background summary for each timestep of each news event. We establish strong baseline performance using state-of-the-art summarization systems and propose a query-focused variant to generate background summaries. To evaluate background summary quality, we present a question-answering-based evaluation metric, Background Utility Score (BUS), which measures the percentage of questions about a current event timestep that a background summary answers. Our experiments show the effectiveness of instruction fine-tuned systems such as Flan-T5, in addition to strong zero-shot performance using GPT-3.5.

Leonardo F. R. Ribeiro, Mengwen Liu, Iryna Gurevych et al. · amazon-science

Despite recent improvements in abstractive summarization, most current approaches generate summaries that are not factually consistent with the source document, severely restricting their trust and usage in real-world applications. Recent works have shown promising improvements in factuality error identification using text or dependency arc entailments; however, they do not consider the entire semantic graph simultaneously. To this end, we propose FactGraph, a method that decomposes the document and the summary into structured meaning representations (MR), which are more suitable for factuality evaluation. MRs describe core semantic concepts and their relations, aggregating the main content in both document and summary in a canonical form, and reducing data sparsity. FactGraph encodes such graphs using a graph encoder augmented with structure-aware adapters to capture interactions among the concepts based on the graph connectivity, along with text representations using an adapter-based text encoder. Experiments on different benchmarks for evaluating factuality show that FactGraph outperforms previous approaches by up to 15%. Furthermore, FactGraph improves performance on identifying content verifiability errors and better captures subsentence-level factual inconsistencies.

Shanshan Zhong, Yi Lu, Jingjie Ning et al.

Skills have become the de facto way to enable LLM agents to perform complex real-world tasks with customized instructions, workflows, and tools, but how to learn them automatically and effectively remains unclear. We introduce SkillLearnBench, the first benchmark for evaluating continual skill learning methods, comprising 20 verified, skill-dependent tasks across 15 sub-domains derived from a real-world skill taxonomy , evaluated at three levels: skill quality, execution trajectory, and task outcome. Using this benchmark, we evaluate recent continual learning techniques, those leveraging one-shot, self/teacher feedback, and skill creator to generate skills from agent experiences. We find that all continual learning methods improve over the no-skill baseline, yet consistent gains remain elusive: no method leads across all tasks and LLMs, and scaling to stronger LLMs does not reliably help. Continual learning improves tasks with clear, reusable workflows but struggles on open-ended tasks, and using stronger LLM backbones does not consistently produce better skills. Our analysis also revealed that multiple iterations in continual learning facilitate genuine improvement via external feedback, whereas self-feedback alone induces recursive drift. Our data and code are open-source at https://github.com/cxcscmu/SkillLearnBench to enable further studies of automatic skill generation and continual learning techniques.

Yukun Huang, Leonardo F. R. Ribeiro, Momchil Hardalov et al.

Search-augmented LLM agents can produce deep research reports (DRRs), but verifying claim-level factuality remains challenging. Existing fact-checkers are primarily designed for general-domain, factoid-style atomic claims, and there is no benchmark to test whether such verifiers transfer to DRRs. Yet building such a benchmark is itself difficult. We first show that static expert-labeled benchmarks are brittle in this setting: in a controlled study with PhD-level specialists, unassisted experts achieve only 60.8% accuracy on a hidden micro-gold set of verifiable claims. We propose Evolving Benchmarking via Audit-then-Score (AtS), where benchmark labels and rationales are explicitly revisable: when a verifier disagrees with the current benchmark, it must submit evidence; an auditor adjudicates the dispute; and accepted revisions update the benchmark before models are scored. Across four AtS rounds, expert micro-gold accuracy rises to 90.9%, indicating experts are substantially more reliable as auditors than as one-shot labelers. We instantiate AtS as DeepFact-Bench, a versioned DRR factuality benchmark with auditable rationales, and DeepFact-Eval, a document-level verification agent (with a grouped lite variant) that outperforms existing verifiers on DeepFact-Bench and transfers well to external factuality datasets.

Prahaladh Chandrahasan, Jiahe Jin, Zhihan Zhang et al.

Effectively evaluating deep research agents that autonomously search the web, analyze information, and generate reports remains a major challenge, particularly when it comes to assessing long reports and giving detailed feedback on their intermediate steps. To address these gaps, we introduce Deep Research Comparator, a platform that offers a holistic framework for deep research agent hosting, side-by-side comparison, fine-grained human feedback collection, and ranking calculation. Given a user query, our platform displays the final reports from two different agents along with their intermediate steps during generation. Annotators can evaluate the overall quality of final reports based on side-by-side comparison, and also provide detailed feedback separately by assessing intermediate steps or specific text spans within the final report. Furthermore, we develop Simple Deepresearch, an end-to-end agent scaffold. This scaffold serves as a baseline that facilitates the easy integration of various large language models to transform them into deep research agents for evaluation. To demonstrate the platform's utility for deep research agent development, we have collected real user preference data from 17 annotators on three deep research agents. A demo video of our platform can be found at https://www.youtube.com/watch?v=g4d2dnbdseg.

Max Glockner, Xiang Jiang, Leonardo F. R. Ribeiro et al.

Evaluating Retrieval-Augmented Generation (RAG) in large language models (LLMs) is challenging because benchmarks can quickly become stale. Questions initially requiring retrieval may become answerable from pretraining knowledge as newer models incorporate more recent information during pretraining, making it difficult to distinguish evidence-based reasoning from recall. We introduce NeoQA (News Events for Out-of-training Question Answering), a benchmark designed to address this issue. To construct NeoQA, we generated timelines and knowledge bases of fictional news events and entities along with news articles and Q\&A pairs to prevent LLMs from leveraging pretraining knowledge, ensuring that no prior evidence exists in their training data. We propose our dataset as a new platform for evaluating evidence-based question answering, as it requires LLMs to generate responses exclusively from retrieved evidence and only when sufficient evidence is available. NeoQA enables controlled evaluation across various evidence scenarios, including cases with missing or misleading details. Our findings indicate that LLMs struggle to distinguish subtle mismatches between questions and evidence, and suffer from short-cut reasoning when key information required to answer a question is missing from the evidence, underscoring key limitations in evidence-based reasoning.

Aashiq Muhamed, Leonardo F. R. Ribeiro, Markus Dreyer et al.

The ability of language models in RAG systems to selectively refuse to answer based on flawed context is critical for safety, yet remains a significant failure point. Our large-scale study reveals that even frontier models struggle in this setting, with refusal accuracy dropping below 50% on multi-document tasks, while exhibiting either dangerous overconfidence or overcaution. Static benchmarks fail to reliably evaluate this capability, as models exploit dataset-specific artifacts and memorize test instances. We introduce RefusalBench, a generative methodology that programmatically creates diagnostic test cases through controlled linguistic perturbation. Our framework employs 176 distinct perturbation strategies across six categories of informational uncertainty and three intensity levels. Evaluation of over 30 models uncovers systematic failure patterns: refusal comprises separable detection and categorization skills, and neither scale nor extended reasoning improves performance. We find that selective refusal is a trainable, alignment-sensitive capability, offering a clear path for improvement. We release two benchmarks -- RefusalBench-NQ (single document) and RefusalBench-GaRAGe (multi-document) -- and our complete generation framework to enable continued, dynamic evaluation of this critical capability.

Alyssa Hwang, Kalpit Dixit, Miguel Ballesteros et al. · amazon-science

We present NewsQs (news-cues), a dataset that provides question-answer pairs for multiple news documents. To create NewsQs, we augment a traditional multi-document summarization dataset with questions automatically generated by a T5-Large model fine-tuned on FAQ-style news articles from the News On the Web corpus. We show that fine-tuning a model with control codes produces questions that are judged acceptable more often than the same model without them as measured through human evaluation. We use a QNLI model with high correlation with human annotations to filter our data. We release our final dataset of high-quality questions, answers, and document clusters as a resource for future work in query-based multi-document summarization.

Markus Dreyer, Mengwen Liu, Feng Nan et al.

Neural models for abstractive summarization tend to generate output that is fluent and well-formed but lacks semantic faithfulness, or factuality, with respect to the input documents. In this paper, we analyze the tradeoff between abstractiveness and factuality of generated summaries across multiple datasets and models, using extensive human evaluations of factuality. In our analysis, we visualize the rates of change in factuality as we gradually increase abstractiveness using a decoding constraint, and we observe that, while increased abstractiveness generally leads to a drop in factuality, the rate of factuality decay depends on factors such as the data that the system was trained on. We introduce two datasets with human factuality judgements; one containing 10.2k generated summaries with systematically varied degrees of abstractiveness; the other containing 4.2k summaries from five different summarization models. We propose new factuality metrics that adjust for the degree of abstractiveness, and we use them to compare the abstractiveness-adjusted factuality of previous summarization works, providing baselines for future work.

Arthur Bražinskas, Mengwen Liu, Ramesh Nallapati et al.

Pre-trained and fine-tuned news summarizers are expected to generalize to news articles unseen in the fine-tuning (training) phase. However, these articles often contain specifics, such as new events and people, a summarizer could not learn about in training. This applies to scenarios such as a news publisher training a summarizer on dated news and summarizing incoming recent news. In this work, we explore the first application of transductive learning to summarization where we further fine-tune models on test set inputs. Specifically, we construct pseudo summaries from salient article sentences and input randomly masked articles. Moreover, this approach is also beneficial in the fine-tuning phase, where we jointly predict extractive pseudo references and abstractive gold summaries in the training set. We show that our approach yields state-of-the-art results on CNN/DM and NYT datasets, improving ROUGE-L by 1.05 and 0.74, respectively. Importantly, our approach does not require any changes of the original architecture. Moreover, we show the benefits of transduction from dated to more recent CNN news. Finally, through human and automatic evaluation, we demonstrate improvements in summary abstractiveness and coherence.

Shiva Pentyala, Mengwen Liu, Markus Dreyer

We present methods for multi-task learning that take advantage of natural groupings of related tasks. Task groups may be defined along known properties of the tasks, such as task domain or language. Such task groups represent supervised information at the inter-task level and can be encoded into the model. We investigate two variants of neural network architectures that accomplish this, learning different feature spaces at the levels of individual tasks, task groups, as well as the universe of all tasks: (1) parallel architectures encode each input simultaneously into feature spaces at different levels; (2) serial architectures encode each input successively into feature spaces at different levels in the task hierarchy. We demonstrate the methods on natural language understanding (NLU) tasks, where a grouping of tasks into different task domains leads to improved performance on ATIS, Snips, and a large inhouse dataset.

Anjishnu Kumar, Arpit Gupta, Julian Chan et al.

This paper presents the design of the machine learning architecture that underlies the Alexa Skills Kit (ASK) a large scale Spoken Language Understanding (SLU) Software Development Kit (SDK) that enables developers to extend the capabilities of Amazon's virtual assistant, Alexa. At Amazon, the infrastructure powers over 25,000 skills deployed through the ASK, as well as AWS's Amazon Lex SLU Service. The ASK emphasizes flexibility, predictability and a rapid iteration cycle for third party developers. It imposes inductive biases that allow it to learn robust SLU models from extremely small and sparse datasets and, in doing so, removes significant barriers to entry for software developers and dialogue systems researchers.

Xing Fan, Emilio Monti, Lambert Mathias et al.

The goal of semantic parsing is to map natural language to a machine interpretable meaning representation language (MRL). One of the constraints that limits full exploration of deep learning technologies for semantic parsing is the lack of sufficient annotation training data. In this paper, we propose using sequence-to-sequence in a multi-task setup for semantic parsing with a focus on transfer learning. We explore three multi-task architectures for sequence-to-sequence modeling and compare their performance with an independently trained model. Our experiments show that the multi-task setup aids transfer learning from an auxiliary task with large labeled data to a target task with smaller labeled data. We see absolute accuracy gains ranging from 1.0% to 4.4% in our in- house data set, and we also see good gains ranging from 2.5% to 7.0% on the ATIS semantic parsing tasks with syntactic and semantic auxiliary tasks.