Despite rapid progress in autonomous web agents, human involvement remains essential for shaping preferences and correcting agent behavior as tasks unfold. However, current agentic systems lack a principled understanding of when and why humans intervene, often proceeding autonomously past critical decision points or requesting unnecessary confirmation. In this work, we introduce the task of modeling human intervention to support collaborative web task execution. We collect CowCorpus, a dataset of 400 real-user web navigation trajectories containing over 4200 interleaved human and agent actions. We identify four distinct patterns of user interaction with agents – hands-off supervision, hands-on oversight, collaborative task-solving, and full user takeover. Leveraging these insights, we train language models (LMs) to anticipate when users are likely to intervene based on their interaction styles, yielding 11.6-18.4% improvement over base LMs. Finally, we deploy these intervention-aware models in live web navigation agents and evaluate them in a user study, finding a 36.8% increase in user-rated agent usefulness. Together, our results show structured modeling of human intervention leads to more adaptive, collaborative agents.

Naturalistic computer-use traces, passively recorded screenshots and mouse or keyboard actions, are a valuable resource for deriving symbolic, auditable, and reusable models of how everyday work is done. Such models matter as computer-use agents enter real work, where agents need to learn how tasks are actually performed, and organizations need to audit and reuse that knowledge. However, inducing such task models is challenging, as activity is observed only as low-level events and real-world work is multi-threaded with interleaved goals. Existing methods assume a given task or single workflow, and produce step-level summaries rather than structured task models. We introduce TASK MODEL INDUCTION (TMI), which (i) discovers the latent tasks in an unconstrained trace, disentangling concurrent activity, and (ii) for each latent task, induces a task model pairing a hierarchical objective model of recursive goal decomposition with a procedure model of the control flow that organized the execution. Intrinsically, on controlled human and agent trajectories, TMI recovers interleaved tasks with 0.974 agreement against ground-truth groupings and reconstructs 74.9% of the observed execution steps, far more than the strongest workflow induction baseline. Extrinsically, skills derived from TMI’s task models improve held-out task accuracy by 30.0% over the strongest baseline.

The recent development of large language models (LLMs) has revolutionized the landscape of human work. These models possess the ability to follow complex human instructions and operate versatile computer software, enabling them to participate in, augment, or even automate realistic occupational tasks that once thought to be exclusive to humans. As LLMs are increasingly integrated into workplaces, they are already reshaping labor dynamics (Hoffmann et al., 2024; Demirci et al., 2025) and raising urgent concerns about job displacement, diminished human agency, and overreliance on automation (Hazra et al., 2025). As a result, the future of work is undergoing a profound transformation: How will human occupations and task requirements evolve over time? And what roles will LLM-based systems play, as they become increasingly capable collaborators and autonomous workers? And how can we build technological and data infrastructures to support human-AI collaboration? This tutorial will provide an overview of the future of work shaped by the interplay of LLMs and humans, examining the emerging challenges, opportunities, and ethical considerations in this dynamic landscape. We begin by examining the economic landscape of work and how NLP technologies drive automation, followed by methods for developing LLMs that augment human labor and recent advances in LLM-based agents. We then cover evaluation approaches for workplace contexts, including datasets, benchmarks, and metrics, and conclude with open questions on technical, human, and societal implications.

The integration of tools has extended the capabilities of language models (LMs) beyond vanilla text generation to versatile scenarios. However, tool-augmented language models (TaLMs) often assume ‘perfect’ information access and tool availability, which may not hold in the real world. To systematically study TaLMs imperfections, we introduce the FAIL-TaLMs benchmark, featuring two major failures: under-specified user queries and non-available tools. FAIL-TaLMS contains 1,749 examples using 906 tools across 21 categories, including single- and multi-tool usage. We evaluate top-performing proprietary and open-source models, and find all current models except for Claude struggle to recognize missing tools or information. Further, to study possible mitigation of the failures, we enable real-time human interaction, named the Ask-and-Help method, to provide missing information or replace non-functional tools. While Ask-and-Help can help models solve tasks more correctly when queries are under-specified, it brings minimal benefit when complex tools are broken.

While much work on web agents emphasizes the promise of autonomously performing tasks on behalf of users, in reality, agents often fallshort on complex tasks in real-world contexts and modeling user preference. This presents an opportunity for humans to collaborate with the agent and leverage the agent’s capabilities effectively. We propose CowPilot, a frame- work supporting autonomous as well as human-agent co llaborative w eb navigation, and evaluation across task success and task efficiency. CowPilot reduces the number of steps humans need to perform by allowing agents to propose next steps, while users are able to pause, reject, or take alternative actions. During execution, users can interleave their actions with the agent’s by overriding suggestions or resuming agent control when needed. We conducted case studies on five common websites and found that the human-agent collaborative mode achieves the highest success rate of 95% while requiring humans to perform only 15.2% of the total steps. Even with human interventions during task execution, the agent successfully drives up to half of task success on its own. CowPilot can serve as a useful tool for data collection and agent evaluation across websites, which we believe will enable research in how users and agents can work together. Video demonstrations are available at https://oaishi.github.io/cowpilot.html

While language models (LMs) excel at generating code, many programs are difficult to generate using only parametric knowledge. Despite the success of retrieval-augmented generation (RAG) in text-centric tasks, its potential for code generation remains under-explored. This work introduces CodeRAG-bench, a holistic retrieval-augmented code generation benchmark covering tasks like basic programming, open-domain, and repository-level problems and provides reproducible evaluations on both retrieval and end-to-end code generation performance. We further create a diverse, open datastore for code retrieval, aggregating sources such as competition solutions, tutorials, library documentation, StackOverflow posts, and GitHub repositories. Based on CodeRAG-bench, we conduct large-scale evaluations of 10 retrievers and 10 LMs and systematically analyze when retrieval can benefit code generation models and identify remaining challenges. We find that while retrieving high-quality contexts improves code generation, retrievers often struggle to fetch useful contexts, and generators face limitations in using those contexts effectively. We hope CodeRAG-bench encourages further development in code-oriented RAG methods.

Retrieval-Augmented Generation (RAG) has become essential for large-scale code generation, grounding predictions in external code corpora to improve factuality. However, a critical yet underexplored aspect of RAG pipelines is chunking—the process of dividing documents into retrievable units. Existing line-based chunking heuristics often break semantic structures, splitting functions or merging unrelated code, which can degrade generation quality. We propose chunking via Abstract Syntax Trees (cAST), a structure-aware method that recursively breaks large AST nodes into smaller chunks and merges sibling nodes while respecting size limits. This approach generates self-contained, semantically coherent units across programming languages and tasks, improving performance on diverse code generation tasks, e.g., boosting Recall@5 by 4.3 points on RepoEval retrieval and Pass@1 by 2.67 points on SWE-bench generation. Our work highlights the importance of structure-aware chunking for scaling retrieval-enhanced code intelligence.

Although large language models (LLMs) have been largely successful in generating functionally correct programs, conditioning models to produce efficient solutions while ensuring correctness remains a challenge. Further, unreliability in benchmarking code efficiency is a hurdle across varying hardware specifications for popular interpreted languages such as Python. In this paper, we present ECCO, a reproducible benchmark for evaluating program efficiency via two paradigms: natural language (NL) based code generation and history-based code editing. On ECCO, we adapt and thoroughly investigate the three most promising existing LLM-based approaches: in-context learning, iterative refinement with execution or NL feedback, and fine-tuning conditioned on execution and editing history. While most methods degrade functional correctness and moderately increase program efficiency, we find that adding execution information often helps maintain functional correctness, and NL feedback enhances more on efficiency. We release our benchmark to support future work on LLM-based generation of efficient code.

Modern abstractive summarization models often generate summaries that contain hallucinated or contradictory information. In this paper, we propose a simple but effective contrastive learning framework that incorporates recent developments in reward learning and factuality metrics. Empirical studies demonstrate that the proposed framework enables summarization models to learn from feedback of factuality metrics using contrastive reward learning, leading to more factual summaries by human evaluations. This suggests that further advances in learning and evaluation algorithms can feed directly into providing more factual summaries. Code and human evaluation results will be publicly available at \url{https://github.com/EthanC111/factuality_summarization}.

To extend the scope of coding queries to more realistic settings, we propose ODEX, the first Open-Domain EXecution-based natural language (NL) to Python code generation dataset. ODEX has 945 NL-Code pairs spanning 79 diverse libraries, along with 1,707 human-written test cases for execution. Our NL-Code pairs are harvested from StackOverflow forums to encourage natural and practical coding queries. Moreover, ODEX supports four natural languages as intents, in English, Spanish, Japanese, and Russian. ODEX unveils intriguing behavioral differences among top-performing code language models (LM). While CODEX achieves better overall results, CODEGEN improves effectively via scaling – CODEGEN 6.1B performs comparably with CODEX 12B. Both models show substantial gaps between open and closed domains, but CODEGEN gaps tend to decrease with model size while CODEX gaps increase. We release ODEX to facilitate research into open-domain problems for the code generation community.

While there has been a recent burgeoning of applications at the intersection of natural and programming languages, such as code generation and code summarization, these applications are usually English-centric. This creates a barrier for program developers who are not proficient in English. To mitigate this gap in technology development across languages, we propose a multilingual dataset, MCoNaLa, to benchmark code generation from natural language commands extending beyond English. Modeled off of the methodology from the English Code/Natural Language Challenge (CoNaLa) dataset, we annotated a total of 896 NL-Code pairs in three languages: Spanish, Japanese, and Russian. We present a systematic evaluation on MCoNaLa by testing state-of-the-art code generation systems. Although the difficulties vary across three languages, all systems lag significantly behind their English counterparts, revealing the challenges in adapting code generation to new languages.

A persistent challenge to table question answering (TableQA) by generating executable programs has been adapting to varied table structures, typically requiring domain-specific logical forms. In response, this paper introduces a unified TableQA framework that: (1) provides a unified representation for structured tables as multi-index Pandas data frames, (2) uses Python as a powerful querying language, and (3) uses few-shot prompting to translate NL questions into Python programs, which are executable on Pandas data frames. Furthermore, to answer complex relational questions with extended program functionality and external knowledge, our framework allows customized APIs that Python programs can call. We experiment with four TableQA datasets that involve tables of different structures — relational, multi-table, and hierarchical matrix shapes — and achieve prominent improvements over past state-of-the-art systems. In ablation studies, we (1) show benefits from our multi-index representation and APIs over baselines that use only an LLM, and (2) demonstrate that our approach is modular and can incorporate additional APIs.

Tables are an important form of structured data for both human and machine readers alike, providing answers to questions that cannot, or cannot easily, be found in texts. Recent work has designed special models and training paradigms for table-related tasks such as table-based question answering and table retrieval. Though effective, they add complexity in both modeling and data acquisition compared to generic text solutions and obscure which elements are truly beneficial. In this work, we focus on the task of table retrieval, and ask: “is table-specific model design necessary for table retrieval, or can a simpler text-based model be effectively used to achieve a similar result?’’ First, we perform an analysis on a table-based portion of the Natural Questions dataset (NQ-table), and find that structure plays a negligible role in more than 70% of the cases. Based on this, we experiment with a general Dense Passage Retriever (DPR) based on text and a specialized Dense Table Retriever (DTR) that uses table-specific model designs. We find that DPR performs well without any table-specific design and training, and even achieves superior results compared to DTR when fine-tuned on properly linearized tables. We then experiment with three modules to explicitly encode table structures, namely auxiliary row/column embeddings, hard attention masks, and soft relation-based attention biases. However, none of these yielded significant improvements, suggesting that table-specific model design may not be necessary for table retrieval.

Systems for knowledge-intensive tasks such as open-domain question answering (QA) usually consist of two stages: efficient retrieval of relevant documents from a large corpus and detailed reading of the selected documents. This is usually done through two separate models, a retriever that encodes the query and finds nearest neighbors, and a reader based on Transformers. These two components are usually modeled separately, which necessitates a cumbersome implementation and is awkward to optimize in an end-to-end fashion. In this paper, we revisit this design and eschew the separate architecture and training in favor of a single Transformer that performs retrieval as attention (RAA), and end-to-end training solely based on supervision from the end QA task. We demonstrate for the first time that an end-to-end trained single Transformer can achieve both competitive retrieval and QA performance on in-domain datasets, matching or even slightly outperforming state-of-the-art dense retrievers and readers. Moreover, end-to-end adaptation of our model significantly boosts its performance on out-of-domain datasets in both supervised and unsupervised settings, making our model a simple and adaptable end-to-end solution for knowledge-intensive tasks.

Tables are often created with hierarchies, but existing works on table reasoning mainly focus on flat tables and neglect hierarchical tables. Hierarchical tables challenge numerical reasoning by complex hierarchical indexing, as well as implicit relationships of calculation and semantics. We present a new dataset, HiTab, to study question answering (QA) and natural language generation (NLG) over hierarchical tables. HiTab is a cross-domain dataset constructed from a wealth of statistical reports and Wikipedia pages, and has unique characteristics: (1) nearly all tables are hierarchical, and (2) QA pairs are not proposed by annotators from scratch, but are revised from real and meaningful sentences authored by analysts. (3) to reveal complex numerical reasoning in statistical reports, we provide fine-grained annotations of quantity and entity alignment. Experiments suggest that this HiTab presents a strong challenge for existing baselines and a valuable benchmark for future research. Targeting hierarchical structure, we devise a hierarchy-aware logical form for symbolic reasoning over tables, which shows high effectiveness. Targeting table reasoning, we leverage entity and quantity alignment to explore partially supervised training in QA and conditional generation in NLG, and largely reduce spurious predictions in QA and produce better descriptions in NLG.

Pre-trained language models like BERT have proven to be highly performant. However, they are often computationally expensive in many practical scenarios, for such heavy models can hardly be readily implemented with limited resources. To improve their efficiency with an assured model performance, we propose a novel speed-tunable FastBERT with adaptive inference time. The speed at inference can be flexibly adjusted under varying demands, while redundant calculation of samples is avoided. Moreover, this model adopts a unique self-distillation mechanism at fine-tuning, further enabling a greater computational efficacy with minimal loss in performance. Our model achieves promising results in twelve English and Chinese datasets. It is able to speed up by a wide range from 1 to 12 times than BERT if given different speedup thresholds to make a speed-performance tradeoff.