Towards Conversational Dataset Retrieval: A Survey

Data Foundations of Long-Context Language Models: A Survey

As the context window of Large Language Models (LLMs) continues to expand, the data required to effectively train and evaluate these capabilities remains underexplored. With existing research primarily focuses on architectural optimization, there is a need for a systematic, data-centric review. This survey bridges this gap by investigating the data foundations of Long-Context Language Models (LCMs). We begin by examining current data strategies alongside their strengths and limitations, mapping the required data to desired model capabilities. Building on this, we explore how targeted training data designs drive core, often interconnected skills such as retrieval, reasoning, and aggregation. Furthermore, we analyze the evaluation landscape, illustrating how selecting appropriate benchmarks is crucial for probing capability boundaries and guiding effective model selection. Finally, we synthesize actionable guidelines for data construction and outline critical future directions to propel the advancement of long-context language models, including quantifying data quality, establishing scaling laws for length distributions, and developing dynamic evaluation frameworks.

Improving Ad-hoc Search Effectiveness for Conversational Information Retrieval via Model Merging

Conversational information retrieval is challenging since it requires the consideration of the conversation history which potentially gives rise to topic shifts and coreference resolution across previous turns. To address these challenges, previous work mainly rely on traditional fine-tuning of ad-hoc retrievers on conversational datasets or extrapolates their generalizability through multi-tasking. However, this mainstream approach is costly—since it requires model re-training—and exhibits catastrophic forgetting, where the model loses its foundational ad-hoc retrieval performance. In this paper, we fill this gap by introducing model merging as a training-free strategy enabling the design of a single retrieval model that operates across both ad-hoc and conversational settings with no additional fine-tuning. We conduct experiments using linear and non-linear parameter-wise merging strategies—namely Model Soup and Slerp—on standard ad-hoc search and conversational retrieval datasets. Our results demonstrate that model merging significantly enhances the ad-hoc search capabilities of conversational retrievers while improving generalizability across task-specific datasets, achieving up to 15% higher NDCG@3 under zero-shot conditions.

APR: Adaptive Personalised Reranking For Conversational Search

Conversational search systems help users satisfy complex information needs through natural language interactions, yet incorporating user preferences into ranking remains challenging. Existing rewrite-then-rerank pipelines capture topical relevance but struggle with fine-grained constraints such as negative preferences or formatting requirements. Instruction-following retrieval approaches are promising for enforcing such constraints, yet their use in personalised conversational search remains underexplored, since ''instructions'' within this context are implicit and situated within user history and profiles, rather than being explicitly stated. We show that instruction-following models can assist with complex queries but introduce noise and latency on simpler keyword queries. To address this issue, we propose Adaptive Personalised Reranking (APR), a framework that routes queries based on intent. APR uses efficient similarity-based reranking for simple queries and dynamically generates tailored instructions to guide an instruction-following reranker for constraint-heavy contexts. Oracle analysis on TREC iKAT 2023 and 2024 shows that instruction-following provides a ''rescue'' potential for hard queries. We also show that APR trained with synthetic data performs competitively against strong baselines such as MonoT5 while offering promising new research avenues.

Report on the 1st Workshop on Conversational Search for Complex Information Needs (CoSCIN’26) at ECIR 2026

The Prompt Is Not the Query: How Request State Evolves Across Multi-Turn AI Conversations

AI-search evaluation commonly treats a prompt as a stable query that can be counted, classified, and replayed in isolation. A conversation makes that unit of analysis questionable: each user turn can add a constraint, revise an assumption, request evidence, or refer to alternatives established earlier. We replace latent"intent"with an observable construct, conversation-conditioned request state, and measure how that state is distributed across user turns. The analysis reuses frozen rules and the governed cohort of a preceding conversation study: 670 English commercial multi-turn conversations in a discovery-replication design and 7,463 public PRISM conversations from 1,389 participants. In the commercial corpus, the final prompt contains a median 35.6% of the session's unique user-side content vocabulary; in PRISM, the median is 36.4%. The final prompt contains at most half of that vocabulary in 68.4% and 74.3% of conversations, respectively. More importantly, transparent rules detect at least one request-state dimension in history but not in the final prompt in 50.3% of commercial conversations and 44.8% of PRISM conversations. Among dimension-bearing conversations, the final prompt reproduces the full observed dimension set in only 26.1% and 26.2%. At the same time, the final prompt adds a previously unseen dimension in 17.9% and 19.3%, showing that the endpoint is neither a summary nor merely a reference: it is often another state update. Length-matched nulls show that low lexical coverage is largely a consequence of turn length, so vocabulary results are interpreted as information availability, not semantic drift. The categorical results support session-level measurement for AI search. They do not estimate the causal effect of history on model answers.

Showdown 2024: Relevance Feedback Meets Conversational Search