Autonomous agents executing human instructions must operate reliably even when instructions are incomplete. While recent approaches improve detection of missing information, detection alone is insufficient: agents often proceed to execution even after recognizing underspecification, leading to incorrect or unsafe actions. We identify this failure as arising from a lack of coupling between detection and execution, and propose that reliable behavior requires enforcing missing information as a precondition for action. We instantiate this principle in Cocoreli, a modular architecture that represents task structure, tracks missing information, and blocks execution until required details are resolved through targeted clarification. In Cocoreli, detection and prevention are structurally coupled: detecting a missing parameter simultaneously blocks execution. We evaluate Cocoreli in a controlled construction environment isolating underspecification and sequential execution. Cocoreli makes execution reliable by imposing explicit task structure: instructions are represented as parameterized executable objects and execution under unresolved specifications is blocked by construction, eliminating hallucinated actions. In contrast, chain-of-thought, prompt-chaining, and ReAct-style reasoning may still execute under incomplete specifications despite high detection rates. The same representation supports abstraction and reuse, and generalizes to API workflow tasks on ToolBench. These results show that reliable collaborative execution under underspecified instructions requires architectural enforcement, not just model capability.

We introduce SagaQA, a long-form video benchmark for multi-hop reasoning over full-length TV series. Existing video reasoning benchmarks often emphasize local understanding of adjacent frames or clips. SagaQA addresses this gap by requiring high-level comprehension of extended multimodal narratives in entire TV shows. A distinguishing feature of SagaQA is the granularity of its reasoning steps. Our dataset necessitates long-range reasoning hops to connect information across completely different episodes. This requires models to reason over entire events and actions, demanding a deep understanding of the show’s narration and progression at a multimodal level. Motivated by recent progress in agentic methods, we further study how different planning strategies handle such complex reasoning. We categorize these approaches into three classes—parallel, sequential, and hybrid planners—and evaluate their ability to generate coherent and complete reasoning plans. Our results on SagaQA suggest that hybrid planners consistently produce higher-quality plans and exhibit stronger capabilities for complex, high-level narrative understanding in TV shows.

Vision-Language Models (VLMs) are able to process increasingly longer videos. Yet, important visual information is easily lost throughout the entire context and missed by VLMs. Also, it is important to design tools that enable cost-effective analysis of lengthy video content. In this paper, we propose a clip selection method that targets key video moments to be included in a multimodal summary. We divide the video into short clips and generate compact visual descriptions of each using a lightweight video captioning model. These are then passed to a large language model (LLM), which selects the K clips containing the most relevant visual information for a multimodal summary. We evaluate our approach on reference clips for the task, automatically derived from full human-annotated screenplays and summaries in the MovieSum dataset. We further show that these reference clips (less than 6% of the movie) are sufficient to build a complete multimodal summary of the movies in MovieSum. Using our clip selection method, we achieve a summarization performance close to that of these reference clips while capturing substantially more relevant video information than random clip selection. Importantly, we maintain low computational cost by relying on a lightweight captioning model.

This paper describes a corpus of situated multiparty chats developed for the STAC project (Strategic Conversation, ERC grant n. 269427). and annotated for discourse structure in the style of Segmented Discourse Representation Theory (SDRT; Asher & Lascarides,2003). The STAC corpus is not only a rich source of data on strategic conversation, but also the first corpus that we are aware of that provides discourse structures for multiparty dialogues situated within a virtual environment. The corpus was annotated in two stages: we initially annotated the chat moves only, but later decided to annotate interactions between the chat moves and non-linguistic events from the virtual environment. This two-step procedure has allowed us quantify various ways in which adding information from the nonlinguistic context affects dialogue structure. In this paper, we look at how annotations based only on linguistic information were preserved once the nonlinguistic context was factored in. We explain that while the preservation of relation instances is relatively high when we move from one corpus to the other, there is little preservation of higher order structures that capture "the main point" of a dialogue and distinguish it from peripheral information.

This paper describes the CASOAR corpus, the first manually annotated corpus that explores the impact of discourse structure on sentiment analysis with a study of movie reviews in French and in English as well as letters to the editor in French. While annotating opinions at the expression, the sentence or the document level is a well-established task and relatively straightforward, discourse annotation remains difficult, especially for non-experts. Therefore, combining both annotations poses several methodological problems that we address here. We propose a multi-layered annotation scheme that includes: the complete discourse structure according to the Segmented Discourse Representation Theory, the opinion orientation of elementary discourse units and opinion expressions, and their associated features. We detail each layer, explore the interactions between them and discuss our results. In particular, we examine the correlation between discourse and semantic category of opinion expressions, the impact of discourse relations on both subjectivity and polarity analysis and the impact of discourse on the determination of the overall opinion of a document. Our results demonstrate that discourse is an important cue for sentiment analysis, at least for the corpus genres we have studied.