Conversational search systems require effective handling of context-dependent queries that often contain ambiguity, omission, and coreference. Conversational Query Reformulation (CQR) addresses this challenge by transforming these queries into self-contained forms suitable for off-the-shelf retrievers. However, existing CQR approaches suffer from two critical constraints: high dependency on costly external supervision from human annotations or large language models, and insufficient alignment between the rewriting model and downstream retrievers. We present ConvSearch-R1, the first self-driven framework that completely eliminates dependency on external rewrite supervision by leveraging reinforcement learning to optimize reformulation directly through retrieval signals. Our novel two-stage approach combines Self-Driven Policy Warm-Up to address the cold-start problem through retrieval-guided self-distillation, followed by Retrieval-Guided Reinforcement Learning with a specially designed rank-incentive reward shaping mechanism that addresses the sparsity issue in conventional retrieval metrics. Extensive experiments on TopiOCQA and QReCC datasets demonstrate that ConvSearch-R1 significantly outperforms previous state-of-the-art methods, achieving over 10% improvement on the challenging TopiOCQA dataset while using smaller 3B parameter models without any external supervision.
Pre-trained language models have been widely exploited to learn dense representations of documents and queries for information retrieval. While previous efforts have primarily focused on improving effectiveness and user satisfaction, response time remains a critical bottleneck of conversational search systems. To address this, we exploit the topical locality inherent in conversational queries, i.e., the tendency of queries within a conversation to focus on related topics. By leveraging query embedding similarities, we dynamically restrict the search space to semantically relevant document clusters, reducing computational complexity without compromising retrieval quality. We evaluate our approach on the TREC CAsT 2019 and 2020 datasets using multiple embedding models and vector indexes, achieving improvements in processing speed of up to 10.4X with little loss in performance (4.4X without any loss). Our results show that the proposed system effectively handles complex, multiturn queries with high precision and efficiency, offering a practical solution for real-time conversational search.
Information retrieval systems have traditionally relied on exact term match methods such as BM25 for first-stage retrieval. However, recent advancements in neural network-based techniques have introduced a new method called dense retrieval. This approach uses a dual-encoder to create contextual embeddings that can be indexed and clustered efficiently at run-time, resulting in improved retrieval performance in Open-domain Question Answering systems. In this paper, we apply the dense retrieval technique to conversational search by conducting experiments on the CAsT benchmark dataset. We also propose an end-to-end conversational search system called GPT2QR+DPR, which incorporates various query reformulation strategies to improve retrieval accuracy. Our findings indicate that dense retrieval outperforms BM25 even without extensive fine-tuning. Our work contributes to the growing body of research on neural-based retrieval methods in conversational search, and highlights the potential of dense retrieval in improving retrieval accuracy in conversational search systems.
The rise of personalized conversational search systems has been driven by advancements in Large Language Models (LLMs), enabling these systems to retrieve and generate answers for complex information needs. However, the automatic evaluation of responses generated by Retrieval Augmented Generation (RAG) systems remains an understudied challenge. In this paper, we introduce a new resource for assessing the retrieval effectiveness and relevance of response generated by RAG systems, using a nugget-based evaluation framework. Built upon the foundation of TREC iKAT 2023, our dataset extends to the TREC iKAT 2024 collection, which includes 17 conversations and 20,575 relevance passage assessments, together with 2,279 extracted gold nuggets, and 62 manually written gold answers from NIST assessors. While maintaining the core structure of its predecessor, this new collection enables a deeper exploration of generation tasks in conversational settings. Key improvements in iKAT 2024 include: (1) ``gold nuggets'' -- concise, essential pieces of information extracted from relevant passages of the collection -- which serve as a foundation for automatic response evaluation; (2) manually written answers to provide a gold standard for response evaluation; (3) unanswerable questions to evaluate model hallucination; (4) expanded user personas, providing richer contextual grounding; and (5) a transition from Personal Text Knowledge Base (PTKB) ranking to PTKB classification and selection. Built on this resource, we provide a framework for long-form answer generation evaluation, involving nuggets extraction and nuggets matching, linked to retrieval. This establishes a solid resource for advancing research in personalized conversational search and long-form answer generation. Our resources are publicly available at https://github.com/irlabamsterdam/CONE-RAG.
Incomplete relevance judgments limit the reusability of test collections. When new systems are compared to previous systems that contributed to the pool, they often face a disadvantage. This is due to pockets of unjudged documents (called holes) in the test collection that the new systems return. The very nature of Conversational Search (CS) means that these holes are potentially larger and more problematic when evaluating systems. In this paper, we aim to extend CS test collections by employing Large Language Models (LLMs) to fill holes by leveraging existing judgments. We explore this problem using TREC iKAT 23 and TREC CAsT 22 collections, where information needs are highly dynamic and the responses are much more varied, leaving bigger holes to fill. Our experiments reveal that CS collections show a trend towards less reusability in deeper turns. Also, fine-tuning the Llama 3.1 model leads to high agreement with human assessors, while few-shot prompting the ChatGPT results in low agreement with humans. Consequently, filling the holes of a new system using ChatGPT leads to a higher change in the location of the new system. While regenerating the assessment pool with few-shot prompting the ChatGPT model and using it for evaluation achieves a high rank correlation with human-assessed pools. We show that filling the holes using few-shot training the Llama 3.1 model enables a fairer comparison between the new system and the systems contributed to the pool. Our hole-filling model based on few-shot training of the Llama 3.1 model can improve the reusability of test collections.
Conversational Search (CS) involves retrieving relevant documents from a corpus while considering the conversational context, integrating retrieval with context modeling. Recent advancements in Large Language Models (LLMs) have significantly enhanced CS by enabling query rewriting based on conversational context. However, employing LLMs during inference poses efficiency challenges. Existing solutions mitigate this issue by distilling embeddings derived from human-rewritten queries, focusing primarily on learning the context modeling task. These methods, however, often separate the contrastive retrieval task from the distillation process, treating it as an independent loss term. To overcome these limitations, we introduce DiSCo (Distillation of Sparse Conversational retrieval), a novel approach that unifies retrieval and context modeling through a relaxed distillation objective. Instead of relying exclusively on representation learning, our method distills similarity scores between conversations and documents, providing more freedom in the representation space and better leveraging the contrastive nature of document relevance. Extensive experiments on Learned Sparse Retrieval (LSR) across five CS datasets demonstrate that DiSCo achieves substantial improvements in both in-domain and out-of-domain retrieval tasks, achieving up to a six-point gain in recall for out-of-domain datasets over state-of-the-art methods. Additionally, DiSCo employs a multi-teacher distillation strategy, using multiple LLMs as teachers, further enhancing performance and surpassing the individual teachers in in-domain settings. Furthermore, analysis of model sparsity reveals that DiSCo allows for more effective control over the sparsity of the trained models.
Conversational Recommender Systems (CRS) proactively engage users in interactive dialogues to elicit user preferences and provide personalized recommendations. Existing methods train Reinforcement Learning (RL)-based agent with greedy action selection or sampling strategy, and may suffer from suboptimal conversational planning. To address this, we present a novel Monte Carlo Tree Search (MCTS)-based CRS framework SAPIENT. SAPIENT consists of a conversational agent (S-agent) and a conversational planner (S-planner). S-planner builds a conversational search tree with MCTS based on the initial actions proposed by S-agent to find conversation plans. The best conversation plans from S-planner are used to guide the training of S-agent, creating a self-training loop where S-agent can iteratively improve its capability for conversational planning. Furthermore, we propose an efficient variant SAPIENT for trade-off between training efficiency and performance. Extensive experiments on four benchmark datasets validate the effectiveness of our approach, showing that SAPIENT outperforms the state-of-the-art baselines. Our code and data are accessible through https://github.com/ninglab/SAPIENT.
Traditional search methods primarily depend on string matches, while semantic search targets concept-based matches by recognizing underlying intents and contextual meanings of search terms. Semantic search is particularly beneficial for discovering scholarly publications where differences in vocabulary between users' search terms and document content are common, often yielding irrelevant search results. Many scholarly search engines have adopted knowledge graphs to represent semantic relations between authors, publications, and research concepts. However, users may face challenges when navigating these graphical search interfaces due to the complexity and volume of data, which impedes their ability to discover publications effectively. To address this problem, we developed a conversational search system for exploring scholarly publications using a knowledge graph. We outline the methodical approach for designing and implementing the proposed system, detailing its architecture and functional components. To assess the system's effectiveness, we employed various performance metrics and conducted a human evaluation with 40 participants, demonstrating how the conversational interface compares against a graphical interface with traditional text search. The findings from our evaluation provide practical insights for advancing the design of conversational search systems.
Conversational search supports multi-turn user-system interactions to solve complex information needs. Different from the traditional single-turn ad-hoc search, conversational search encounters a more challenging problem of context-dependent query understanding with the lengthy and long-tail conversational history context. While conversational query rewriting methods leverage explicit rewritten queries to train a rewriting model to transform the context-dependent query into a stand-stone search query, this is usually done without considering the quality of search results. Conversational dense retrieval methods use fine-tuning to improve a pre-trained ad-hoc query encoder, but they are limited by the conversational search data available for training. In this paper, we leverage both rewritten queries and relevance judgments in the conversational search data to train a better query representation model. The key idea is to align the query representation with those of rewritten queries and relevant documents. The proposed model -- Query Representation Alignment Conversational Dense Retriever, QRACDR, is tested on eight datasets, including various settings in conversational search and ad-hoc search. The results demonstrate the strong performance of QRACDR compared with state-of-the-art methods, and confirm the effectiveness of representation alignment.
With large language models (LLMs), conversational search engines shift how users retrieve information from the web by enabling natural conversations to express their search intents over multiple turns. Users' natural conversation embodies rich but implicit signals of users' search intents and evaluation of search results to understand user experience with the system. However, it is underexplored how and why users ask follow-up queries to continue conversations with conversational search engines and how the follow-up queries signal users' satisfaction. From qualitative analysis of 250 conversational turns from an in-lab user evaluation of Naver Cue:, a commercial conversational search engine, we propose a taxonomy of 18 users' follow-up query patterns from conversational search, comprising two major axes: (1) users' motivations behind continuing conversations (N = 7) and (2) actions of follow-up queries (N = 11). Compared to the existing literature on query reformulations, we uncovered a new set of motivations and actions behind follow-up queries, including asking for subjective opinions or providing natural language feedback on the engine's responses. To analyze conversational search logs with our taxonomy in a scalable and efficient manner, we built an LLM-powered classifier (73% accuracy). With our classifier, we analyzed 2,061 conversational tuples collected from real-world usage logs of Cue: and examined how the conversation patterns from our taxonomy correlates with satisfaction. Our initial findings suggest some signals of dissatisfactions, such as Clarifying Queries, Excluding Condition, and Substituting Condition with follow-up queries. We envision our approach could contribute to automated evaluation of conversation search experience by providing satisfaction signals and grounds for realistic user simulations.
Conversational Query Reformulation (CQR) has significantly advanced in addressing the challenges of conversational search, particularly those stemming from the latent user intent and the need for historical context. Recent works aimed to boost the performance of CQR through alignment. However, they are designed for one specific retrieval system, which potentially results in sub-optimal generalization. To overcome this limitation, we present a novel framework AdaCQR. By aligning reformulation models with both term-based and semantic-based retrieval systems, AdaCQR enhances the generalizability of information-seeking queries among diverse retrieval environments through a two-stage training strategy. Moreover, two effective approaches are proposed to obtain superior labels and diverse input candidates, boosting the efficiency and robustness of the framework. Experimental results on the TopiOCQA and QReCC datasets demonstrate that AdaCQR outperforms the existing methods in a more efficient framework, offering both quantitative and qualitative improvements in conversational query reformulation.
In this paper, we study how open-source large language models (LLMs) can be effectively deployed for improving query rewriting in conversational search, especially for ambiguous queries. We introduce CHIQ, a two-step method that leverages the capabilities of LLMs to resolve ambiguities in the conversation history before query rewriting. This approach contrasts with prior studies that predominantly use closed-source LLMs to directly generate search queries from conversation history. We demonstrate on five well-established benchmarks that CHIQ leads to state-of-the-art results across most settings, showing highly competitive performances with systems leveraging closed-source LLMs. Our study provides a first step towards leveraging open-source LLMs in conversational search, as a competitive alternative to the prevailing reliance on commercial LLMs. Data, models, and source code will be publicly available upon acceptance at https://github.com/fengranMark/CHIQ.
Major search engine providers are rapidly incorporating Large Language Model (LLM)-generated content in response to user queries. These conversational search engines operate by loading retrieved website text into the LLM context for summarization and interpretation. Recent research demonstrates that LLMs are highly vulnerable to jailbreaking and prompt injection attacks, which disrupt the safety and quality goals of LLMs using adversarial strings. This work investigates the impact of prompt injections on the ranking order of sources referenced by conversational search engines. To this end, we introduce a focused dataset of real-world consumer product websites and formalize conversational search ranking as an adversarial problem. Experimentally, we analyze conversational search rankings in the absence of adversarial injections and show that different LLMs vary significantly in prioritizing product name, document content, and context position. We then present a tree-of-attacks-based jailbreaking technique which reliably promotes low-ranked products. Importantly, these attacks transfer effectively to state-of-the-art conversational search engines such as perplexity$.$ai. Given the strong financial incentive for website owners to boost their search ranking, we argue that our problem formulation is of critical importance for future robustness work.
Incomplete relevance judgments limit the re-usability of test collections. When new systems are compared against previous systems used to build the pool of judged documents, they often do so at a disadvantage due to the ``holes'' in test collection (i.e., pockets of un-assessed documents returned by the new system). In this paper, we take initial steps towards extending existing test collections by employing Large Language Models (LLM) to fill the holes by leveraging and grounding the method using existing human judgments. We explore this problem in the context of Conversational Search using TREC iKAT, where information needs are highly dynamic and the responses (and, the results retrieved) are much more varied (leaving bigger holes). While previous work has shown that automatic judgments from LLMs result in highly correlated rankings, we find substantially lower correlates when human plus automatic judgments are used (regardless of LLM, one/two/few shot, or fine-tuned). We further find that, depending on the LLM employed, new runs will be highly favored (or penalized), and this effect is magnified proportionally to the size of the holes. Instead, one should generate the LLM annotations on the whole document pool to achieve more consistent rankings with human-generated labels. Future work is required to prompt engineering and fine-tuning LLMs to reflect and represent the human annotations, in order to ground and align the models, such that they are more fit for purpose.
The future of conversational agents will provide users with personalized information responses. However, a significant challenge in developing models is the lack of large-scale dialogue datasets that span multiple sessions and reflect real-world user preferences. Previous approaches rely on experts in a wizard-of-oz setup that is difficult to scale, particularly for personalized tasks. Our method, LAPS, addresses this by using large language models (LLMs) to guide a single human worker in generating personalized dialogues. This method has proven to speed up the creation process and improve quality. LAPS can collect large-scale, human-written, multi-session, and multi-domain conversations, including extracting user preferences. When compared to existing datasets, LAPS-produced conversations are as natural and diverse as expert-created ones, which stays in contrast with fully synthetic methods. The collected dataset is suited to train preference extraction and personalized response generation. Our results show that responses generated explicitly using extracted preferences better match user's actual preferences, highlighting the value of using extracted preferences over simple dialogue history. Overall, LAPS introduces a new method to leverage LLMs to create realistic personalized conversational data more efficiently and effectively than previous methods.
The increasing reliance on digital information necessitates advancements in conversational search systems, particularly in terms of information transparency. While prior research in conversational information-seeking has concentrated on improving retrieval techniques, the challenge remains in generating responses useful from a user perspective. This study explores different methods of explaining the responses, hypothesizing that transparency about the source of the information, system confidence, and limitations can enhance users' ability to objectively assess the response. By exploring transparency across explanation type, quality, and presentation mode, this research aims to bridge the gap between system-generated responses and responses verifiable by the user. We design a user study to answer questions concerning the impact of (1) the quality of explanations enhancing the response on its usefulness and (2) ways of presenting explanations to users. The analysis of the collected data reveals lower user ratings for noisy explanations, although these scores seem insensitive to the quality of the response. Inconclusive results on the explanations presentation format suggest that it may not be a critical factor in this setting.
Conversational information seeking has evolved rapidly in the last few years with the development of Large Language Models (LLMs), providing the basis for interpreting and responding in a naturalistic manner to user requests. The extended TREC Interactive Knowledge Assistance Track (iKAT) collection aims to enable researchers to test and evaluate their Conversational Search Agents (CSA). The collection contains a set of 36 personalized dialogues over 20 different topics each coupled with a Personal Text Knowledge Base (PTKB) that defines the bespoke user personas. A total of 344 turns with approximately 26,000 passages are provided as assessments on relevance, as well as additional assessments on generated responses over four key dimensions: relevance, completeness, groundedness, and naturalness. The collection challenges CSA to efficiently navigate diverse personal contexts, elicit pertinent persona information, and employ context for relevant conversations. The integration of a PTKB and the emphasis on decisional search tasks contribute to the uniqueness of this test collection, making it an essential benchmark for advancing research in conversational and interactive knowledge assistants.
Conversational search requires accurate interpretation of user intent from complex multi-turn contexts. This paper presents ChatRetriever, which inherits the strong generalization capability of large language models to robustly represent complex conversational sessions for dense retrieval. To achieve this, we propose a simple and effective dual-learning approach that adapts LLM for retrieval via contrastive learning while enhancing the complex session understanding through masked instruction tuning on high-quality conversational instruction tuning data. Extensive experiments on five conversational search benchmarks demonstrate that ChatRetriever substantially outperforms existing conversational dense retrievers, achieving state-of-the-art performance on par with LLM-based rewriting approaches. Furthermore, ChatRetriever exhibits superior robustness in handling diverse conversational contexts. Our work highlights the potential of adapting LLMs for retrieval with complex inputs like conversational search sessions and proposes an effective approach to advance this research direction.
Conversational search provides a more convenient interface for users to search by allowing multi-turn interaction with the search engine. However, the effectiveness of the conversational dense retrieval methods is limited by the scarcity of training data required for their fine-tuning. Thus, generating more training conversational sessions with relevant labels could potentially improve search performance. Based on the promising capabilities of large language models (LLMs) on text generation, we propose ConvSDG, a simple yet effective framework to explore the feasibility of boosting conversational search by using LLM for session data generation. Within this framework, we design dialogue/session-level and query-level data generation with unsupervised and semi-supervised learning, according to the availability of relevance judgments. The generated data are used to fine-tune the conversational dense retriever. Extensive experiments on four widely used datasets demonstrate the effectiveness and broad applicability of our ConvSDG framework compared with several strong baselines.
Conversational information-seeking (CIS) is an emerging paradigm for knowledge acquisition and exploratory search. Traditional web search interfaces enable easy exploration of entities, but this is limited in conversational settings due to the limited-bandwidth interface. This paper explore ways to rewrite answers in CIS, so that users can understand them without having to resort to external services or sources. Specifically, we focus on salient entities -- entities that are central to understanding the answer. As our first contribution, we create a dataset of conversations annotated with entities for saliency. Our analysis of the collected data reveals that the majority of answers contain salient entities. As our second contribution, we propose two answer rewriting strategies aimed at improving the overall user experience in CIS. One approach expands answers with inline definitions of salient entities, making the answer self-contained. The other approach complements answers with follow-up questions, offering users the possibility to learn more about specific entities. Results of a crowdsourcing-based study indicate that rewritten answers are clearly preferred over the original ones. We also find that inline definitions tend to be favored over follow-up questions, but this choice is highly subjective, thereby providing a promising future direction for personalization.