Avishek Anand

Mandeep Rathee, Thorben Funke, Avishek Anand et al.

The problem of interpreting the decisions of machine learning is a well-researched and important. We are interested in a specific type of machine learning model that deals with graph data called graph neural networks. Evaluating interpretability approaches for graph neural networks (GNN) specifically are known to be challenging due to the lack of a commonly accepted benchmark. Given a GNN model, several interpretability approaches exist to explain GNN models with diverse (sometimes conflicting) evaluation methodologies. In this paper, we propose a benchmark for evaluating the explainability approaches for GNNs called Bagel. In Bagel, we firstly propose four diverse GNN explanation evaluation regimes -- 1) faithfulness, 2) sparsity, 3) correctness. and 4) plausibility. We reconcile multiple evaluation metrics in the existing literature and cover diverse notions for a holistic evaluation. Our graph datasets range from citation networks, document graphs, to graphs from molecules and proteins. We conduct an extensive empirical study on four GNN models and nine post-hoc explanation approaches for node and graph classification tasks. We open both the benchmarks and reference implementations and make them available at https://github.com/Mandeep-Rathee/Bagel-benchmark.

Yumeng Wang, Lijun Lyu, Avishek Anand

Contextual ranking models based on BERT are now well established for a wide range of passage and document ranking tasks. However, the robustness of BERT-based ranking models under adversarial inputs is under-explored. In this paper, we argue that BERT-rankers are not immune to adversarial attacks targeting retrieved documents given a query. Firstly, we propose algorithms for adversarial perturbation of both highly relevant and non-relevant documents using gradient-based optimization methods. The aim of our algorithms is to add/replace a small number of tokens to a highly relevant or non-relevant document to cause a large rank demotion or promotion. Our experiments show that a small number of tokens can already result in a large change in the rank of a document. Moreover, we find that BERT-rankers heavily rely on the document start/head for relevance prediction, making the initial part of the document more susceptible to adversarial attacks. More interestingly, we find a small set of recurring adversarial words that when added to documents result in successful rank demotion/promotion of any relevant/non-relevant document respectively. Finally, our adversarial tokens also show particular topic preferences within and across datasets, exposing potential biases from BERT pre-training or downstream datasets.

Abhijit Anand, Venktesh V, Vinay Setty et al.

Query rewriting refers to an established family of approaches that are applied to underspecified and ambiguous queries to overcome the vocabulary mismatch problem in document ranking. Queries are typically rewritten during query processing time for better query modelling for the downstream ranker. With the advent of large-language models (LLMs), there have been initial investigations into using generative approaches to generate pseudo documents to tackle this inherent vocabulary gap. In this work, we analyze the utility of LLMs for improved query rewriting for text ranking tasks. We find that there are two inherent limitations of using LLMs as query re-writers -- concept drift when using only queries as prompts and large inference costs during query processing. We adopt a simple, yet surprisingly effective, approach called context aware query rewriting (CAR) to leverage the benefits of LLMs for query understanding. Firstly, we rewrite ambiguous training queries by context-aware prompting of LLMs, where we use only relevant documents as context.Unlike existing approaches, we use LLM-based query rewriting only during the training phase. Eventually, a ranker is fine-tuned on the rewritten queries instead of the original queries during training. In our extensive experiments, we find that fine-tuning a ranker using re-written queries offers a significant improvement of up to 33% on the passage ranking task and up to 28% on the document ranking task when compared to the baseline performance of using original queries.

Avishek Anand, Venktesh V, Abhijit Anand et al.

Querying, conversing, and controlling search and information-seeking interfaces using natural language are fast becoming ubiquitous with the rise and adoption of large-language models (LLM). In this position paper, we describe a generic framework for interactive query-rewriting using LLMs. Our proposal aims to unfold new opportunities for improved and transparent intent understanding while building high-performance retrieval systems using LLMs. A key aspect of our framework is the ability of the rewriter to fully specify the machine intent by the search engine in natural language that can be further refined, controlled, and edited before the final retrieval phase. The ability to present, interact, and reason over the underlying machine intent in natural language has profound implications on transparency, ranking performance, and a departure from the traditional way in which supervised signals were collected for understanding intents. We detail the concept, backed by initial experiments, along with open questions for this interactive query understanding framework.

Simone Piaggesi, Megha Khosla, André Panisson et al.

Graphs are ubiquitous due to their flexibility in representing social and technological systems as networks of interacting elements. Graph representation learning methods, such as node embeddings, are powerful approaches to map nodes into a latent vector space, allowing their use for various graph tasks. Despite their success, only few studies have focused on explaining node embeddings locally. Moreover, global explanations of node embeddings remain unexplored, limiting interpretability and debugging potentials. We address this gap by developing human-understandable explanations for dimensions in node embeddings. Towards that, we first develop new metrics that measure the global interpretability of embedding vectors based on the marginal contribution of the embedding dimensions to predicting graph structure. We say that an embedding dimension is more interpretable if it can faithfully map to an understandable sub-structure in the input graph - like community structure. Having observed that standard node embeddings have low interpretability, we then introduce DINE (Dimension-based Interpretable Node Embedding), a novel approach that can retrofit existing node embeddings by making them more interpretable without sacrificing their task performance. We conduct extensive experiments on synthetic and real-world graphs and show that we can simultaneously learn highly interpretable node embeddings with effective performance in link prediction.

Julia Maria Struß, Sebastian Schellhammer, Stefan Dietze et al.

The CheckThat! lab aims to advance the development of innovative technologies combating disinformation and manipulation efforts in online communication across a multitude of languages and platforms. While in early editions the focus has been on core tasks of the verification pipeline (check-worthiness, evidence retrieval, and verification), in the past three editions, the lab added additional tasks linked to the verification process. In this year's edition, the verification pipeline is at the center again with the following tasks: Task 1 on source retrieval for scientific web claims (a follow-up of the 2025 edition), Task 2 on fact-checking numerical and temporal claims, which adds a reasoning component to the 2025 edition, and Task 3, which expands the verification pipeline with generation of full-fact-checking articles. These tasks represent challenging classification and retrieval problems as well as generation challenges at the document and span level, including multilingual settings.

Bhawna Piryani, Abdelrahman Abdallah, Jamshid Mozafari et al.

Time plays a critical role in how information is generated, retrieved, and interpreted. In this survey, we provide a comprehensive overview of Temporal Question Answering (TQA), a research area that focuses on answering questions involving temporal constraints or context. As time-stamped content from sources like news articles, web archives, and knowledge bases continues to grow, TQA systems must address challenges such as detecting temporal intent, normalizing time expressions, ordering events, and reasoning over evolving or ambiguous facts. We organize existing work through a unified perspective that captures the interaction between corpus temporality, question temporality, and model capabilities, enabling a systematic comparison of datasets, tasks, and approaches. We review recent advances in TQA enabled by neural architectures, especially transformer-based models and Large Language Models (LLMs), highlighting progress in temporal language modeling, retrieval-augmented generation (RAG), and temporal reasoning. We also discuss benchmark datasets and evaluation strategies designed to test temporal robustness,

Zijian Zhang, Vinay Setty, Avishek Anand

We introduce SparcAssist, a general-purpose risk assessment tool for the machine learning models trained for language tasks. It evaluates models' risk by inspecting their behavior on counterfactuals, namely out-of-distribution instances generated based on the given data instance. The counterfactuals are generated by replacing tokens in rational subsequences identified by ExPred, while the replacements are retrieved using HotFlip or Masked-Language-Model-based algorithms. The main purpose of our system is to help the human annotators to assess the model's risk on deployment. The counterfactual instances generated during the assessment are the by-product and can be used to train more robust NLP models in the future.

Harrie Oosterhuis, Lijun Lyu, Avishek Anand

Local feature selection in machine learning provides instance-specific explanations by focusing on the most relevant features for each prediction, enhancing the interpretability of complex models. However, such methods tend to produce misleading explanations by encoding additional information in their selections. In this work, we attribute the problem of misleading selections by formalizing the concepts of label and feature leakage. We rigorously derive the necessary and sufficient conditions under which we can guarantee no leakage, and show existing methods do not meet these conditions. Furthermore, we propose the first local feature selection method that is proven to have no leakage called SUWR. Our experimental results indicate that SUWR is less prone to overfitting and combines state-of-the-art predictive performance with high feature-selection sparsity. Our generic and easily extendable formal approach provides a strong theoretical basis for future work on interpretability with reliable explanations.

Jonas Wallat, Tianyi Zhang, Avishek Anand

Language models retain a significant amount of world knowledge from their pre-training stage. This allows knowledgeable models to be applied to knowledge-intensive tasks prevalent in information retrieval, such as ranking or question answering. Understanding how and which factual information is acquired by our models is necessary to build responsible models. However, limited work has been done to understand the effect of pre-training tasks on the amount of knowledge captured and forgotten by language models during pre-training. Building a better understanding of knowledge acquisition is the goal of this paper. Therefore, we utilize a selection of pre-training tasks to infuse knowledge into our model. In the following steps, we test the model's knowledge retention by measuring its ability to answer factual questions. Our experiments show that masking entities and principled masking of correlated spans based on pointwise mutual information lead to more factual knowledge being retained than masking random tokens. Our findings demonstrate that, like the ability to perform a task, the (factual) knowledge acquired from being trained on that task is forgotten when a model is trained to perform another task (catastrophic forgetting) and how to prevent this phenomenon. To foster reproducibility, the code, as well as the data used in this paper, are openly available.

Venktesh V, Sourangshu Bhattacharya, Avishek Anand

Answering complex questions is a challenging task that requires question decomposition and multistep reasoning for arriving at the solution. While existing supervised and unsupervised approaches are specialized to a certain task and involve training, recently proposed prompt-based approaches offer generalizable solutions to tackle a wide variety of complex question-answering (QA) tasks. However, existing prompt-based approaches that are effective for complex QA tasks involve expensive hand annotations from experts in the form of rationales and are not generalizable to newer complex QA scenarios and tasks. We propose, icat (In-Context Ability Transfer) which induces reasoning capabilities in LLMs without any LLM fine-tuning or manual annotation of in-context samples. We transfer the ability to decompose complex questions to simpler questions or generate step-by-step rationales to LLMs, by careful selection from available data sources of related tasks. We also propose an automated uncertainty-aware exemplar selection approach for selecting examples from transfer data sources. Finally, we conduct large-scale experiments on a variety of complex QA tasks involving numerical reasoning, compositional complex QA, and heterogeneous complex QA which require decomposed reasoning. We show that ICAT convincingly outperforms existing prompt-based solutions without involving any model training, showcasing the benefits of re-using existing abilities.

Abhijit Anand, Jurek Leonhardt, V Venktesh et al.

As information retrieval systems continue to evolve, accurate evaluation and benchmarking of these systems become pivotal. Web search datasets, such as MS MARCO, primarily provide short keyword queries without accompanying intent or descriptions, posing a challenge in comprehending the underlying information need. This paper proposes an approach to augmenting such datasets to annotate informative query descriptions, with a focus on two prominent benchmark datasets: TREC-DL-21 and TREC-DL-22. Our methodology involves utilizing state-of-the-art LLMs to analyze and comprehend the implicit intent within individual queries from benchmark datasets. By extracting key semantic elements, we construct detailed and contextually rich descriptions for these queries. To validate the generated query descriptions, we employ crowdsourcing as a reliable means of obtaining diverse human perspectives on the accuracy and informativeness of the descriptions. This information can be used as an evaluation set for tasks such as ranking, query rewriting, or others.

Jonas Wallat, Adam Jatowt, Avishek Anand

Large language models (LLMs) have recently gained significant attention due to their unparalleled ability to perform various natural language processing tasks. These models, benefiting from their advanced natural language understanding capabilities, have demonstrated impressive zero-shot performance. However, the pre-training data utilized in LLMs is often confined to a specific corpus, resulting in inherent freshness and temporal scope limitations. Consequently, this raises concerns regarding the effectiveness of LLMs for tasks involving temporal intents. In this study, we aim to investigate the underlying limitations of general-purpose LLMs when deployed for tasks that require a temporal understanding. We pay particular attention to handling factual temporal knowledge through three popular temporal QA datasets. Specifically, we observe low performance on detailed questions about the past and, surprisingly, for rather new information. In manual and automatic testing, we find multiple temporal errors and characterize the conditions under which QA performance deteriorates. Our analysis contributes to understanding LLM limitations and offers valuable insights into developing future models that can better cater to the demands of temporally-oriented tasks. The code is available\footnote{https://github.com/jwallat/temporalblindspots}.

Kiran Purohit, Venktesh V, Raghuram Devalla et al.

Answering reasoning-based complex questions over text and hybrid sources, including tables, is a challenging task. Recent advances in large language models (LLMs) have enabled in-context learning (ICL), allowing LLMs to acquire proficiency in a specific task using only a few demonstration samples (exemplars). A critical challenge in ICL is the selection of optimal exemplars, which can be either task-specific (static) or test-example-specific (dynamic). Static exemplars provide faster inference times and increased robustness across a distribution of test examples. In this paper, we propose an algorithm for static exemplar subset selection for complex reasoning tasks. We introduce EXPLORA, a novel exploration method designed to estimate the parameters of the scoring function, which evaluates exemplar subsets without incorporating confidence information. EXPLORA significantly reduces the number of LLM calls to ~11% of those required by state-of-the-art methods and achieves a substantial performance improvement of 12.24%. We open-source our code and data (https://github.com/kiranpurohit/EXPLORA).

V Venktesh, Mandeep Rathee, Avishek Anand

Test-time scaling (TTS) has emerged as a new frontier for scaling the performance of Large Language Models. In test-time scaling, by using more computational resources during inference, LLMs can improve their reasoning process and task performance. Several approaches have emerged for TTS such as distilling reasoning traces from another model or exploring the vast decoding search space by employing a verifier. The verifiers serve as reward models that help score the candidate outputs from the decoding process to diligently explore the vast solution space and select the best outcome. This paradigm commonly termed has emerged as a superior approach owing to parameter free scaling at inference time and high performance gains. The verifiers could be prompt-based, fine-tuned as a discriminative or generative model to verify process paths, outcomes or both. Despite their widespread adoption, there is no detailed collection, clear categorization and discussion of diverse verification approaches and their training mechanisms. In this survey, we cover the diverse approaches in the literature and present a unified view of verifier training, types and their utility in test-time scaling. Our repository can be found at https://github.com/elixir-research-group/Verifierstesttimescaling.github.io.

Kiran Purohit, V Venktesh, Sourangshu Bhattacharya et al.

The in-context learning paradigm with LLMs has been instrumental in advancing a wide range of natural language processing tasks. The selection of few-shot examples (exemplars / demonstration samples) is essential for constructing effective prompts under context-length budget constraints. In this paper, we formulate the exemplar selection task as a top-m best arms identification problem. A key challenge in this setup is the exponentially large number of arms that need to be evaluated to identify the m-best arms. We propose CASE (Challenger Arm Sampling for Exemplar selection), a novel sample-efficient selective exploration strategy that maintains a shortlist of "challenger" arms, which are current candidates for the top-m arms. In each iteration, only one of the arms from this shortlist or the current topm set is pulled, thereby reducing sample complexity and, consequently, the number of LLM evaluations. Furthermore, we model the scores of exemplar subsets (arms) using a parameterized linear scoring function, leading to stochastic linear bandits setting. CASE achieves remarkable efficiency gains of up to 7x speedup in runtime while requiring 7x fewer LLM calls (87% reduction) without sacrificing performance compared to state-of-the-art exemplar selection methods. We release our code and data at https://github.com/kiranpurohit/CASE

V Venktesh, Deepali Prabhu, Avishek Anand

Fact-checking numerical claims is critical as the presence of numbers provide mirage of veracity despite being fake potentially causing catastrophic impacts on society. The prior works in automatic fact verification do not primarily focus on natural numerical claims. A typical human fact-checker first retrieves relevant evidence addressing the different numerical aspects of the claim and then reasons about them to predict the veracity of the claim. Hence, the search process of a human fact-checker is a crucial skill that forms the foundation of the verification process. Emulating a real-world setting is essential to aid in the development of automated methods that encompass such skills. However, existing benchmarks employ heuristic claim decomposition approaches augmented with weakly supervised web search to collect evidences for verifying claims. This sometimes results in less relevant evidences and noisy sources with temporal leakage rendering a less realistic retrieval setting for claim verification. Hence, we introduce QuanTemp++: a dataset consisting of natural numerical claims, an open domain corpus, with the corresponding relevant evidence for each claim. The evidences are collected through a claim decomposition process approximately emulating the approach of human fact-checker and veracity labels ensuring there is no temporal leakage. Given this dataset, we also characterize the retrieval performance of key claim decomposition paradigms. Finally, we observe their effect on the outcome of the verification pipeline and draw insights. The code for data pipeline along with link to data can be found at https://github.com/VenkteshV/QuanTemp_Plus

Primakov Chungkham, V Venktesh, Vinay Setty et al.

Fact-checking real-world claims, particularly numerical claims, is inherently complex that require multistep reasoning and numerical reasoning for verifying diverse aspects of the claim. Although large language models (LLMs) including reasoning models have made tremendous advances, they still fall short on fact-checking real-world claims that require a combination of compositional and numerical reasoning. They are unable to understand nuance of numerical aspects, and are also susceptible to the reasoning drift issue, where the model is unable to contextualize diverse information resulting in misinterpretation and backtracking of reasoning process. In this work, we systematically explore scaling test-time compute (TTS) for LLMs on the task of fact-checking complex numerical claims, which entails eliciting multiple reasoning paths from an LLM. We train a verifier model (VERIFIERFC) to navigate this space of possible reasoning paths and select one that could lead to the correct verdict. We observe that TTS helps mitigate the reasoning drift issue, leading to significant performance gains for fact-checking numerical claims. To improve compute efficiency in TTS, we introduce an adaptive mechanism that performs TTS selectively based on the perceived complexity of the claim. This approach achieves 1.8x higher efficiency than standard TTS, while delivering a notable 18.8% performance improvement over single-shot claim verification methods. Our code and data can be found at https://github.com/VenkteshV/VerifierFC

Alexandru Dumitru, V Venktesh, Adam Jatowt et al.

Large Language Models (LLMs) have demonstrated immense advances in a wide range of natural language tasks. However, these models are susceptible to hallucinations and errors on particularly temporal understanding tasks involving multiple entities in answers. In such tasks, they fail to associate entities with accurate time intervals, generate a complete list of entities in answers or reason about events associated with specific temporal bounds. Existing works do not extensively evaluate the abilities of the model to perform implicit and explicit temporal understanding in a list answer construction setup. To bridge this gap, we propose the Time referenced List based Question Answering or TLQA benchmark that requires structured answers in list format aligned with corresponding time periods. Our TLQA benchmark, requires both list construction and temporal understanding simultaneously, which to the best of our knowledge has not been explored in prior benchmarks. We investigate the temporal understanding and list construction capabilities of state-of-the-art generative models on TLQA in closed-book and open-domain settings. Our findings reveal significant shortcomings in current models, particularly their inability to provide complete answers and temporally align facts in a closed-book setup and the need to improve retrieval in open-domain setup, providing clear future directions for research on TLQA. The benchmark and code at https://github.com/elixir-research-group/TLQA.

Venktesh V. Deepali Prabhu, Avishek Anand

Open-domain complex Question Answering (QA) is a difficult task with challenges in evidence retrieval and reasoning. The complexity of such questions could stem from questions being compositional, hybrid evidence, or ambiguity in questions. While retrieval performance for classical QA tasks is well explored, their capabilities for heterogeneous complex retrieval tasks, especially in an open-domain setting, and the impact on downstream QA performance, are relatively unexplored. To address this, in this work, we propose a benchmark composing diverse complex QA tasks and provide a toolkit to evaluate state-of-the-art pre-trained dense and sparse retrieval models in an open-domain setting. We observe that late interaction models and surprisingly lexical models like BM25 perform well compared to other pre-trained dense retrieval models. In addition, since context-based reasoning is critical for solving complex QA tasks, we also evaluate the reasoning capabilities of LLMs and the impact of retrieval performance on their reasoning capabilities. Through experiments, we observe that much progress is to be made in retrieval for complex QA to improve downstream QA performance. Our software and related data can be accessed at https://github.com/VenkteshV/DEXTER

Venktesh V, Abhijit Anand, Avishek Anand et al.

Automated fact checking has gained immense interest to tackle the growing misinformation in the digital era. Existing systems primarily focus on synthetic claims on Wikipedia, and noteworthy progress has also been made on real-world claims. In this work, we release QuanTemp, a diverse, multi-domain dataset focused exclusively on numerical claims, encompassing temporal, statistical and diverse aspects with fine-grained metadata and an evidence collection without leakage. This addresses the challenge of verifying real-world numerical claims, which are complex and often lack precise information, not addressed by existing works that mainly focus on synthetic claims. We evaluate and quantify the limitations of existing solutions for the task of verifying numerical claims. We also evaluate claim decomposition based methods, numerical understanding based models and our best baselines achieves a macro-F1 of 58.32. This demonstrates that QuanTemp serves as a challenging evaluation set for numerical claim verification.

V Venktesh, Mandeep Rathee, Avishek Anand

Modern retrieval pipelines increasingly rely on query reformulation and neural reranking to improve effectiveness, but this comes at a significant computational cost and introduces a fundamental tradeoff between recall and query drift. Generating many reformulated queries can substantially increase recall, yet naively merging or exhaustively reranking their results is prohibitively expensive. In this work, we argue that the core challenge is not reformulation generation itself, but the adaptive selection of reformulations and their retrieved documents under a strict inference budget. We propose ReformIR, a budget-aware retrieval framework that treats query reformulations as first-class features and performs online relevance estimation using a strong reranker as a teacher. Given multiple reformulated queries, ReformIR constructs a large candidate pool and learns a lightweight surrogate model that estimates document utility from reformulation-specific retrieval signals. Under a fixed reranking budget, the surrogate adaptively prioritizes both reformulations and documents, selectively querying a teacher reranker anchored to the original query. This process increases recall while actively suppressing drift through online feature selection over reformulations. We conduct extensive experiments on the MSMARCO passage corpora and TREC Deep Learning benchmarks (DL19-DL22). Our results show that ReformIR consistently outperforms existing reformulation strategies, particularly as the number of reformulations increases, where prior methods suffer from severe quality degradation due to drift. Our findings also suggest a shift in retrieval system design, rather than using large language models as rerankers, their capacity is more effectively leveraged in the reformulation stage with feedback-driven optimization.

Mandeep Rathee, V Venktesh, Sean MacAvaney et al.

The classical cascading pipeline of retrieve--rerank suffers from a bounded recall problem, stemming from limitations of the first-stage retriever. Most current approaches address the bounded recall problem by improving the first-stage retriever, but this incurs substantial training and inference costs, especially to handle queries that require substantial reasoning. To circumvent the computational costs of reasoning-based retrievers, we replicate the findings of GAR, Graph-based Adaptive Reranking, on the BRIGHT reasoning-intensive retrieval benchmark. GAR addresses the bounded recall problem by modifying the reranking process itself through iterative exploration of a corpus graph, but it was previously only tested on models designed for topical and question-answering-style queries. Hence, reproduce GAR in reasoning-intensive settings with reasoning and non-reasoning reranking models. We observe that the quality of the reranker's signal plays an important role in identifying additional relevant documents within the corpus graph. Overall, we find that GAR boosts the effectiveness of reasoning-intensive retrieval across a variety of models while contributing minimally to computational overheads. Ultimately, this work enables more practical deployment of retrieval systems that can address reasoning-intensive queries.

Mandeep Rathee, Sean MacAvaney, Avishek Anand

Large Language Models (LLMs) have shown strong promise as rerankers, especially in ``listwise'' settings where an LLM is prompted to rerank several search results at once. However, this ``cascading'' retrieve-and-rerank approach is limited by the bounded recall problem: relevant documents not retrieved initially are permanently excluded from the final ranking. Adaptive retrieval techniques address this problem, but do not work with listwise rerankers because they assume a document's score is computed independently from other documents. In this paper, we propose an adaptation of an existing adaptive retrieval method that supports the listwise setting and helps guide the retrieval process itself (thereby overcoming the bounded recall problem for LLM rerankers). Specifically, our proposed algorithm merges results both from the initial ranking and feedback documents provided by the most relevant documents seen up to that point. Through extensive experiments across diverse LLM rerankers, first stage retrievers, and feedback sources, we demonstrate that our method can improve nDCG@10 by up to 13.23% and recall by 28.02%--all while keeping the total number of LLM inferences constant and overheads due to the adaptive process minimal. The work opens the door to leveraging LLM-based search in settings where the initial pool of results is limited, e.g., by legacy systems, or by the cost of deploying a semantic first-stage.

Jonas Wallat, Abdelrahman Abdallah, Adam Jatowt et al.

Large Language Models (LLMs) encapsulate a surprising amount of factual world knowledge. However, their performance on temporal questions and historical knowledge is limited because they often cannot understand temporal scope and orientation or neglect the temporal aspect altogether. In this study, we aim to measure precisely how robust LLMs are for question answering based on their ability to process temporal information and perform tasks requiring temporal reasoning and temporal factual knowledge. Specifically, we design eight time-sensitive robustness tests for factual information to check the sensitivity of six popular LLMs in the zero-shot setting. Overall, we find LLMs lacking temporal robustness, especially to temporal reformulations and the use of different granularities of temporal references. We show how a selection of these eight tests can be used automatically to judge a model's temporal robustness for user questions on the fly. Finally, we apply the findings of this study to improve the temporal QA performance by up to 55 percent.

Abdelrahman Abdallah, Bhawna Piryani, Jonas Wallat et al.

Temporal awareness is crucial in many information retrieval tasks, particularly in scenarios where the relevance of documents depends on their alignment with the query's temporal context. Traditional approaches such as BM25 and Dense Passage Retrieval (DPR) focus on lexical or semantic similarity but tend to neglect the temporal alignment between queries and documents, which is essential for time-sensitive tasks like temporal question answering (TQA). We propose TempRetriever, a novel extension of DPR that explicitly incorporates temporal information by embedding both the query date and document timestamp into the retrieval process. This allows retrieving passages that are not only contextually relevant but also aligned with the temporal intent of queries. We evaluate TempRetriever on two large-scale datasets ArchivalQA and ChroniclingAmericaQA demonstrating its superiority over baseline retrieval models across multiple metrics. TempRetriever achieves a 6.63\% improvement in Top-1 retrieval accuracy and a 3.79\% improvement in NDCG@10 compared to the standard DPR on ArchivalQA. Similarly, for ChroniclingAmericaQA, TempRetriever exhibits a 9.56\% improvement in Top-1 retrieval accuracy and a 4.68\% improvement in NDCG@10. We also propose a novel, time-based negative sampling strategy which further enhances retrieval performance by addressing temporal misalignment during training. Our results underline the importance of temporal aspects in dense retrieval systems and establish a new benchmark for time-aware passage retrieval.

Jonas Wallat, Maria Heuss, Maarten de Rijke et al.

Retrieving relevant context is a common approach to reduce hallucinations and enhance answer reliability. Explicitly citing source documents allows users to verify generated responses and increases trust. Prior work largely evaluates citation correctness - whether cited documents support the corresponding statements. But citation correctness alone is insufficient. To establish trust in attributed answers, we must examine both citation correctness and citation faithfulness. In this work, we first disentangle the notions of citation correctness and faithfulness, which have been applied inconsistently in previous studies. Faithfulness ensures that the model's reliance on cited documents is genuine, reflecting actual reference use rather than superficial alignment with prior beliefs, which we call post-rationalization. We design an experiment that reveals the prevalent issue of post-rationalization, which undermines reliable attribution and may result in misplaced trust. Our findings suggest that current attributed answers often lack citation faithfulness (up to 57 percent of the citations), highlighting the need to evaluate correctness and faithfulness for trustworthy attribution in language models.

Mandeep Rathee, V Venktesh, Sean MacAvaney et al.

Retrieval-Augmented Generation (RAG) has emerged as a standard framework for knowledge-intensive NLP tasks, combining large language models (LLMs) with document retrieval from external corpora. Despite its widespread use, most RAG pipelines continue to treat retrieval and reasoning as isolated components, retrieving documents once and then generating answers without further interaction. This static design often limits performance on complex tasks that require iterative evidence gathering or high-precision retrieval. Recent work in both the information retrieval (IR) and NLP communities has begun to close this gap by introducing adaptive retrieval and ranking methods that incorporate feedback. In this survey, we present a structured overview of advanced retrieval and ranking mechanisms that integrate such feedback. We categorize feedback signals based on their source and role in improving the query, retrieved context, or document pool. By consolidating these developments, we aim to bridge IR and NLP perspectives and highlight retrieval as a dynamic, learnable component of end-to-end RAG systems.

Lea Richtmann, Viktoria-S. Schmiesing, Dennis Wilken et al.

Setting up and controlling optical systems is often a challenging and tedious task. The high number of degrees of freedom to control mirrors, lenses, or phases of light makes automatic control challenging, especially when the complexity of the system cannot be adequately modeled due to noise or non-linearities. Here, we show that reinforcement learning (RL) can overcome these challenges when coupling laser light into an optical fiber, using a model-free RL approach that trains directly on the experiment without pre-training on simulations. By utilizing the sample-efficient algorithms Soft Actor-Critic (SAC), Truncated Quantile Critics (TQC), or CrossQ, our agents learn to couple with 90% efficiency. A human expert reaches this efficiency, but the RL agents are quicker. In particular, the CrossQ agent outperforms the other agents in coupling speed while requiring only half the training time. We demonstrate that direct training on an experiment can replace extensive system modeling. Our result exemplifies RL's potential to tackle problems in optics, paving the way for more complex applications where full noise modeling is not feasible.

Christine Bauer, Li Chen, Nicola Ferro et al.

During the workshop, we deeply discussed what CONversational Information ACcess (CONIAC) is and its unique features, proposing a world model abstracting it, and defined the Conversational Agents Framework for Evaluation (CAFE) for the evaluation of CONIAC systems, consisting of six major components: 1) goals of the system's stakeholders, 2) user tasks to be studied in the evaluation, 3) aspects of the users carrying out the tasks, 4) evaluation criteria to be considered, 5) evaluation methodology to be applied, and 6) measures for the quantitative criteria chosen.

Amir Abolfazli, Zekun Song, Avishek Anand et al.

The success of deep reinforcement learning (DRL) relies on the availability and quality of training data, often requiring extensive interactions with specific environments. In many real-world scenarios, where data collection is costly and risky, offline reinforcement learning (RL) offers a solution by utilizing data collected by domain experts and searching for a batch-constrained optimal policy. This approach is further augmented by incorporating external data sources, expanding the range and diversity of data collection possibilities. However, existing offline RL methods often struggle with challenges posed by non-matching data from these external sources. In this work, we specifically address the problem of source-target domain mismatch in scenarios involving mixed datasets, characterized by a predominance of source data generated from random or suboptimal policies and a limited amount of target data generated from higher-quality policies. To tackle this problem, we introduce Transition Scoring (TS), a novel method that assigns scores to transitions based on their similarity to the target domain, and propose Curriculum Learning-Based Trajectory Valuation (CLTV), which effectively leverages these transition scores to identify and prioritize high-quality trajectories through a curriculum learning approach. Our extensive experiments across various offline RL methods and MuJoCo environments, complemented by rigorous theoretical analysis, demonstrate that CLTV enhances the overall performance and transferability of policies learned by offline RL algorithms.

Zijian Zhang, Vinay Setty, Yumeng Wang et al.

With the rapid advancement of neural language models, the deployment of over-parameterized models has surged, increasing the need for interpretable explanations comprehensible to human inspectors. Existing post-hoc interpretability methods, which often focus on unigram features of single input textual instances, fail to capture the models' decision-making process fully. Additionally, many methods do not differentiate between decisions based on spurious correlations and those based on a holistic understanding of the input. Our paper introduces DISCO, a novel method for discovering global, rule-based explanations by identifying causal n-gram associations with model predictions. This method employs a scalable sequence mining technique to extract relevant text spans from training data, associate them with model predictions, and conduct causality checks to distill robust rules that elucidate model behavior. These rules expose potential overfitting and provide insights into misleading feature combinations. We validate DISCO through extensive testing, demonstrating its superiority over existing methods in offering comprehensive insights into complex model behaviors. Our approach successfully identifies all shortcuts manually introduced into the training data (100% detection rate on the MultiRC dataset), resulting in an 18.8% regression in model performance -- a capability unmatched by any other method. Furthermore, DISCO supports interactive explanations, enabling human inspectors to distinguish spurious causes in the rule-based output. This alleviates the burden of abundant instance-wise explanations and helps assess the model's risk when encountering out-of-distribution (OOD) data.

Abhijit Anand, Venktesh V, Vinay Setty et al.

An important problem in text-ranking systems is handling the hard queries that form the tail end of the query distribution. The difficulty may arise due to the presence of uncommon, underspecified, or incomplete queries. In this work, we improve the ranking performance of hard or difficult queries without compromising the performance of other queries. Firstly, we do LLM based query enrichment for training queries using relevant documents. Next, a specialized ranker is fine-tuned only on the enriched hard queries instead of the original queries. We combine the relevance scores from the specialized ranker and the base ranker, along with a query performance score estimated for each query. Our approach departs from existing methods that usually employ a single ranker for all queries, which is biased towards easy queries, which form the majority of the query distribution. In our extensive experiments on the DL-Hard dataset, we find that a principled query performance based scoring method using base and specialized ranker offers a significant improvement of up to 25% on the passage ranking task and up to 48.4% on the document ranking task when compared to the baseline performance of using original queries, even outperforming SOTA model.

Jurek Leonhardt, Koustav Rudra, Megha Khosla et al.

Neural document ranking approaches, specifically transformer models, have achieved impressive gains in ranking performance. However, query processing using such over-parameterized models is both resource and time intensive. In this paper, we propose the Fast-Forward index -- a simple vector forward index that facilitates ranking documents using interpolation of lexical and semantic scores -- as a replacement for contextual re-rankers and dense indexes based on nearest neighbor search. Fast-Forward indexes rely on efficient sparse models for retrieval and merely look up pre-computed dense transformer-based vector representations of documents and passages in constant time for fast CPU-based semantic similarity computation during query processing. We propose index pruning and theoretically grounded early stopping techniques to improve the query processing throughput. We conduct extensive large-scale experiments on TREC-DL datasets and show improvements over hybrid indexes in performance and query processing efficiency using only CPUs. Fast-Forward indexes can provide superior ranking performance using interpolation due to the complementary benefits of lexical and semantic similarities.

Zijian Zhang, Koustav Rudra, Avishek Anand

Fact-checking on the Web has become the main mechanism through which we detect the credibility of the news or information. Existing fact-checkers verify the authenticity of the information (support or refute the claim) based on secondary sources of information. However, existing approaches do not consider the problem of model updates due to constantly increasing training data due to user feedback. It is therefore important to conduct user studies to correct models' inference biases and improve the model in a life-long learning manner in the future according to the user feedback. In this paper, we present FaxPlainAC, a tool that gathers user feedback on the output of explainable fact-checking models. FaxPlainAC outputs both the model decision, i.e., whether the input fact is true or not, along with the supporting/refuting evidence considered by the model. Additionally, FaxPlainAC allows for accepting user feedback both on the prediction and explanation. Developed in Python, FaxPlainAC is designed as a modular and easily deployable tool. It can be integrated with other downstream tasks and allowing for fact-checking human annotation gathering and life-long learning.

Mandeep Rathee, Zijian Zhang, Thorben Funke et al.

Graph neural networks (GNNs) have achieved great success on various tasks and fields that require relational modeling. GNNs aggregate node features using the graph structure as inductive biases resulting in flexible and powerful models. However, GNNs remain hard to interpret as the interplay between node features and graph structure is only implicitly learned. In this paper, we propose a novel method called Kedge for explicitly sparsifying the underlying graph by removing unnecessary neighbors. Our key idea is based on a tractable method for sparsification using the Hard Kumaraswamy distribution that can be used in conjugation with any GNN model. Kedge learns edge masks in a modular fashion trained with any GNN allowing for gradient based optimization in an end-to-end fashion. We demonstrate through extensive experiments that our model Kedge can prune a large proportion of the edges with only a minor effect on the test accuracy. Specifically, in the PubMed dataset, Kedge learns to drop more than 80% of the edges with an accuracy drop of merely 2% showing that graph structure has only a small contribution in comparison to node features. Finally, we also show that Kedge effectively counters the over-smoothing phenomena in deep GNNs by maintaining good task performance with increasing GNN layers.

Jurek Leonhardt, Koustav Rudra, Avishek Anand

Neural document ranking models perform impressively well due to superior language understanding gained from pre-training tasks. However, due to their complexity and large number of parameters, these (typically transformer-based) models are often non-interpretable in that ranking decisions can not be clearly attributed to specific parts of the input documents. In this paper we propose ranking models that are inherently interpretable by generating explanations as a by-product of the prediction decision. We introduce the Select-and-Rank paradigm for document ranking, where we first output an explanation as a selected subset of sentences in a document. Thereafter, we solely use the explanation or selection to make the prediction, making explanations first-class citizens in the ranking process. Technically, we treat sentence selection as a latent variable trained jointly with the ranker from the final output. To that end, we propose an end-to-end training technique for Select-and-Rank models utilizing reparameterizable subset sampling using the Gumbel-max trick. We conduct extensive experiments to demonstrate that our approach is competitive to state-of-the-art methods. Our approach is broadly applicable to numerous ranking tasks and furthers the goal of building models that are interpretable by design. Finally, we present real-world applications that benefit from our sentence selection method.

Michael Völske, Alexander Bondarenko, Maik Fröbe et al.

Recently, neural networks have been successfully employed to improve upon state-of-the-art performance in ad-hoc retrieval tasks via machine-learned ranking functions. While neural retrieval models grow in complexity and impact, little is understood about their correspondence with well-studied IR principles. Recent work on interpretability in machine learning has provided tools and techniques to understand neural models in general, yet there has been little progress towards explaining ranking models. We investigate whether one can explain the behavior of neural ranking models in terms of their congruence with well understood principles of document ranking by using established theories from axiomatic IR. Axiomatic analysis of information retrieval models has formalized a set of constraints on ranking decisions that reasonable retrieval models should fulfill. We operationalize this axiomatic thinking to reproduce rankings based on combinations of elementary constraints. This allows us to investigate to what extent the ranking decisions of neural rankers can be explained in terms of retrieval axioms, and which axioms apply in which situations. Our experimental study considers a comprehensive set of axioms over several representative neural rankers. While the existing axioms can already explain the particularly confident ranking decisions rather well, future work should extend the axiom set to also cover the other still "unexplainable" neural IR rank decisions.

Jurek Leonhardt, Fabian Beringer, Avishek Anand

Recently, pre-trained contextual models, such as BERT, have shown to perform well in language related tasks. We revisit the design decisions that govern the applicability of these models for the passage re-ranking task in open-domain question answering. We find that common approaches in the literature rely on fine-tuning a pre-trained BERT model and using a single, global representation of the input, discarding useful fine-grained relevance signals in token- or sentence-level representations. We argue that these discarded tokens hold useful information that can be leveraged. In this paper, we explicitly model the sentence-level representations by using Dynamic Memory Networks (DMNs) and conduct empirical evaluation to show improvements in passage re-ranking over fine-tuned vanilla BERT models by memory-enhanced explicit sentence modelling on a diverse set of open-domain QA datasets. We further show that freezing the BERT model and only training the DMN layer still comes close to the original performance, while improving training efficiency drastically. This indicates that the usual fine-tuning step mostly helps to aggregate the inherent information in a single output token, as opposed to adapting the whole model to the new task, and only achieves rather small gains.

Jonas Wallat, Jaspreet Singh, Avishek Anand

Probing complex language models has recently revealed several insights into linguistic and semantic patterns found in the learned representations. In this article, we probe BERT specifically to understand and measure the relational knowledge it captures in its parametric memory. While probing for linguistic understanding is commonly applied to all layers of BERT as well as fine-tuned models, this has not been done for factual knowledge. We utilize existing knowledge base completion tasks (LAMA) to probe every layer of pre-trained as well as fine-tuned BERT models(ranking, question answering, NER). Our findings show that knowledge is not just contained in BERT's final layers. Intermediate layers contribute a significant amount (17-60%) to the total knowledge found. Probing intermediate layers also reveals how different types of knowledge emerge at varying rates. When BERT is fine-tuned, relational knowledge is forgotten. The extent of forgetting is impacted by the fine-tuning objective and the training data. We found that ranking models forget the least and retain more knowledge in their final layer compared to masked language modeling and question-answering. However, masked language modeling performed the best at acquiring new knowledge from the training data. When it comes to learning facts, we found that capacity and fact density are key factors. We hope this initial work will spur further research into understanding the parametric memory of language models and the effect of training objectives on factual knowledge. The code to repeat the experiments is publicly available on GitHub.

Thorben Funke, Megha Khosla, Mandeep Rathee et al.

With the ever-increasing popularity and applications of graph neural networks, several proposals have been made to explain and understand the decisions of a graph neural network. Explanations for graph neural networks differ in principle from other input settings. It is important to attribute the decision to input features and other related instances connected by the graph structure. We find that the previous explanation generation approaches that maximize the mutual information between the label distribution produced by the model and the explanation to be restrictive. Specifically, existing approaches do not enforce explanations to be valid, sparse, or robust to input perturbations. In this paper, we lay down some of the fundamental principles that an explanation method for graph neural networks should follow and introduce a metric RDT-Fidelity as a measure of the explanation's effectiveness. We propose a novel approach Zorro based on the principles from rate-distortion theory that uses a simple combinatorial procedure to optimize for RDT-Fidelity. Extensive experiments on real and synthetic datasets reveal that Zorro produces sparser, stable, and more faithful explanations than existing graph neural network explanation approaches.

Maximilian Idahl, Lijun Lyu, Ujwal Gadiraju et al.

Post-hoc explanation methods are an important class of approaches that help understand the rationale underlying a trained model's decision. But how useful are they for an end-user towards accomplishing a given task? In this vision paper, we argue the need for a benchmark to facilitate evaluations of the utility of post-hoc explanation methods. As a first step to this end, we enumerate desirable properties that such a benchmark should possess for the task of debugging text classifiers. Additionally, we highlight that such a benchmark facilitates not only assessing the effectiveness of explanations but also their efficiency.

Koustav Rudra, Zeon Trevor Fernando, Avishek Anand

Pre-trained contextual language models such as BERT, GPT, and XLnet work quite well for document retrieval tasks. Such models are fine-tuned based on the query-document/query-passage level relevance labels to capture the ranking signals. However, the documents are longer than the passages and such document ranking models suffer from the token limitation (512) of BERT. Researchers proposed ranking strategies that either truncate the documents beyond the token limit or chunk the documents into units that can fit into the BERT. In the later case, the relevance labels are either directly transferred from the original query-document pair or learned through some external model. In this paper, we conduct a detailed study of the design decisions about splitting and label transfer on retrieval effectiveness and efficiency. We find that direct transfer of relevance labels from documents to passages introduces label noise that strongly affects retrieval effectiveness for large training datasets. We also find that query processing times are adversely affected by fine-grained splitting schemes. As a remedy, we propose a careful passage level labelling scheme using weak supervision that delivers improved performance (3-14% in terms of nDCG score) over most of the recently proposed models for ad-hoc retrieval while maintaining manageable computational complexity on four diverse document retrieval datasets.

Zijian Zhang, Jaspreet Singh, Ujwal Gadiraju et al.

Complex machine learning models are deployed in several critical domains including healthcare and autonomous vehicles nowadays, albeit as functional black boxes. Consequently, there has been a recent surge in interpreting decisions of such complex models in order to explain their actions to humans. Models that correspond to human interpretation of a task are more desirable in certain contexts and can help attribute liability, build trust, expose biases and in turn build better models. It is, therefore, crucial to understand how and which models conform to human understanding of tasks. In this paper, we present a large-scale crowdsourcing study that reveals and quantifies the dissonance between human and machine understanding, through the lens of an image classification task. In particular, we seek to answer the following questions: Which (well-performing) complex ML models are closer to humans in their use of features to make accurate predictions? How does task difficulty affect the feature selection capability of machines in comparison to humans? Are humans consistently better at selecting features that make image recognition more accurate? Our findings have important implications on human-machine collaboration, considering that a long term goal in the field of artificial intelligence is to make machines capable of learning and reasoning like humans.

Zijian Zhang, Koustav Rudra, Avishek Anand

A desirable property of learning systems is to be both effective and interpretable. Towards this goal, recent models have been proposed that first generate an extractive explanation from the input text and then generate a prediction on just the explanation called explain-then-predict models. These models primarily consider the task input as a supervision signal in learning an extractive explanation and do not effectively integrate rationales data as an additional inductive bias to improve task performance. We propose a novel yet simple approach ExPred, that uses multi-task learning in the explanation generation phase effectively trading-off explanation and prediction losses. And then we use another prediction network on just the extracted explanations for optimizing the task performance. We conduct an extensive evaluation of our approach on three diverse language datasets -- fact verification, sentiment classification, and QA -- and find that we substantially outperform existing approaches.

Jonas Wallat, Jaspreet Singh, Avishek Anand

Probing complex language models has recently revealed several insights into linguistic and semantic patterns found in the learned representations. In this paper, we probe BERT specifically to understand and measure the relational knowledge it captures. We utilize knowledge base completion tasks to probe every layer of pre-trained as well as fine-tuned BERT (ranking, question answering, NER). Our findings show that knowledge is not just contained in BERT's final layers. Intermediate layers contribute a significant amount (17-60%) to the total knowledge found. Probing intermediate layers also reveals how different types of knowledge emerge at varying rates. When BERT is fine-tuned, relational knowledge is forgotten but the extent of forgetting is impacted by the fine-tuning objective but not the size of the dataset. We found that ranking models forget the least and retain more knowledge in their final layer. We release our code on github to repeat the experiments.

Avishek Anand, Lawrence Cavedon, Matthias Hagen et al.

Dagstuhl Seminar 19461 "Conversational Search" was held on 10-15 November 2019. 44~researchers in Information Retrieval and Web Search, Natural Language Processing, Human Computer Interaction, and Dialogue Systems were invited to share the latest development in the area of Conversational Search and discuss its research agenda and future directions. A 5-day program of the seminar consisted of six introductory and background sessions, three visionary talk sessions, one industry talk session, and seven working groups and reporting sessions. The seminar also had three social events during the program. This report provides the executive summary, overview of invited talks, and findings from the seven working groups which cover the definition, evaluation, modelling, explanation, scenarios, applications, and prototype of Conversational Search. The ideas and findings presented in this report should serve as one of the main sources for diverse research programs on Conversational Search.

Jaspreet Singh, Zhenye Wang, Megha Khosla et al.

Learning-to-rank (LTR) is a class of supervised learning techniques that apply to ranking problems dealing with a large number of features. The popularity and widespread application of LTR models in prioritizing information in a variety of domains makes their scrutability vital in today's landscape of fair and transparent learning systems. However, limited work exists that deals with interpreting the decisions of learning systems that output rankings. In this paper we propose a model agnostic local explanation method that seeks to identify a small subset of input features as explanation to a ranking decision. We introduce new notions of validity and completeness of explanations specifically for rankings, based on the presence or absence of selected features, as a way of measuring goodness. We devise a novel optimization problem to maximize validity directly and propose greedy algorithms as solutions. In extensive quantitative experiments we show that our approach outperforms other model agnostic explanation approaches across pointwise, pairwise and listwise LTR models in validity while not compromising on completeness.

Rishiraj Saha Roy, Avishek Anand

The last few years have seen an explosion of research on the topic of automated question answering (QA), spanning the communities of information retrieval, natural language processing, and artificial intelligence. This tutorial would cover the highlights of this really active period of growth for QA to give the audience a grasp over the families of algorithms that are currently being used. We partition research contributions by the underlying source from where answers are retrieved: curated knowledge graphs, unstructured text, or hybrid corpora. We choose this dimension of partitioning as it is the most discriminative when it comes to algorithm design. Other key dimensions are covered within each sub-topic: like the complexity of questions addressed, and degrees of explainability and interactivity introduced in the systems. We would conclude the tutorial with the most promising emerging trends in the expanse of QA, that would help new entrants into this field make the best decisions to take the community forward. Much has changed in the community since the last tutorial on QA in SIGIR 2016, and we believe that this timely overview will indeed benefit a large number of conference participants.

Jurek Leonhardt, Avishek Anand, Megha Khosla

The extraction of main content from web pages is an important task for numerous applications, ranging from usability aspects, like reader views for news articles in web browsers, to information retrieval or natural language processing. Existing approaches are lacking as they rely on large amounts of hand-crafted features for classification. This results in models that are tailored to a specific distribution of web pages, e.g. from a certain time frame, but lack in generalization power. We propose a neural sequence labeling model that does not rely on any hand-crafted features but takes only the HTML tags and words that appear in a web page as input. This allows us to present a browser extension which highlights the content of arbitrary web pages directly within the browser using our model. In addition, we create a new, more current dataset to show that our model is able to adapt to changes in the structure of web pages and outperform the state-of-the-art model.