Debabrata Mahapatra

Debabrata Mahapatra, Chaosheng Dong, Yetian Chen et al.

Learning to Rank (LTR) technique is ubiquitous in the Information Retrieval system nowadays, especially in the Search Ranking application. The query-item relevance labels typically used to train the ranking model are often noisy measurements of human behavior, e.g., product rating for product search. The coarse measurements make the ground truth ranking non-unique with respect to a single relevance criterion. To resolve ambiguity, it is desirable to train a model using many relevance criteria, giving rise to Multi-Label LTR (MLLTR). Moreover, it formulates multiple goals that may be conflicting yet important to optimize for simultaneously, e.g., in product search, a ranking model can be trained based on product quality and purchase likelihood to increase revenue. In this research, we leverage the Multi-Objective Optimization (MOO) aspect of the MLLTR problem and employ recently developed MOO algorithms to solve it. Specifically, we propose a general framework where the information from labels can be combined in a variety of ways to meaningfully characterize the trade-off among the goals. Our framework allows for any gradient based MOO algorithm to be used for solving the MLLTR problem. We test the proposed framework on two publicly available LTR datasets and one e-commerce dataset to show its efficacy.

Sarthak Harne, Natwar Modani, Debabrata Mahapatra et al.

Conversational product search assistants offer a more expressive, natural, and interactive alternative to traditional keyword-based product search. With limited screen space, showing only a few items increases the need for precise preference elicitation, which can prolong conversations, leading to user frustration and session abandonment. Conversely, rushing to recommend items without a clear understanding of preferences risks poor matches and a degraded user experience. We present Dialogue to Discovery (D2D), an attribute-oriented preference elicitation framework that dynamically exploits the structure of product attributes to efficiently steer conversations toward the user's desired item. D2D adaptively prioritizes the most informative queries and strategically times product recommendations, reducing premature or off-target suggestions that harm engagement. To evaluate D2D, we curate three datasets from the Amazon Reviews corpus. In simulated conversations modelled using a multi-factor utilitarian patience framework, D2D achieves a 22.2-29.9% improvement in target-finding accuracy, 6.6-16.1% reduction in abandonment, and 27.5% shorter average conversations over the state-of-the-art baselines. A complementary user study further confirms significant gains in both user satisfaction and perceived efficiency.

Shubham Agarwal, Sai Sundaresan, Subrata Mitra et al.

Retrieval-Augmented Generation (RAG) is often used with Large Language Models (LLMs) to infuse domain knowledge or user-specific information. In RAG, given a user query, a retriever extracts chunks of relevant text from a knowledge base. These chunks are sent to an LLM as part of the input prompt. Typically, any given chunk is repeatedly retrieved across user questions. However, currently, for every question, attention-layers in LLMs fully compute the key values (KVs) repeatedly for the input chunks, as state-of-the-art methods cannot reuse KV-caches when chunks appear at arbitrary locations with arbitrary contexts. Naive reuse leads to output quality degradation. This leads to potentially redundant computations on expensive GPUs and increases latency. In this work, we propose Cache-Craft, a system for managing and reusing precomputed KVs corresponding to the text chunks (we call chunk-caches) in RAG-based systems. We present how to identify chunk-caches that are reusable, how to efficiently perform a small fraction of recomputation to fix the cache to maintain output quality, and how to efficiently store and evict chunk-caches in the hardware for maximizing reuse while masking any overheads. With real production workloads as well as synthetic datasets, we show that Cache-Craft reduces redundant computation by 51% over SOTA prefix-caching and 75% over full recomputation. Additionally, with continuous batching on a real production workload, we get a 1.6X speed up in throughput and a 2X reduction in end-to-end response latency over prefix-caching while maintaining quality, for both the LLaMA-3-8B and LLaMA-3-70B models.

Archit Gupta, Sai Sundaresan, Debabrata Mahapatra

As enterprises deploy RAG-based systems to provide grounded responses to user queries, reranking has become a critical component for the final filtering step that separates relevant from distracting or irrelevant documents. Existing rerankers often rely on heuristic thresholds to achieve optimal filtering. Moreover, for relevance scoring, state-of-the-art methods use a language model's logit signals, which are designed for next-token prediction, not for assessing relevance. To address these limitations, we identify a principled signal for relevance: the representational shift (RS) induced in a query's internal state when conditioned on a document. We observe that the alignment between (a) RS induced by a candidate document and (b) RS induced by an oracle document-set provides a robust indicator of relevance. Building on this insight, we introduce a lightweight training framework that learns projections mapping RS to calibrated relevance scores. Our training objectives naturally filter irrelevant content at a zero threshold, reducing dependence on heuristic tuning. Across diverse retrieval datasets, our method delivers gains over SOTA rerankers.

Debabrata Mahapatra, Shubham Agarwal, Apoorv Saxena et al.

Prior work on input-token importance in auto-regressive transformers has relied on Softmax-normalized attention weights, which obscure the richer structure of pre-Softmax query-key logits. We introduce RCStat, a statistical framework that harnesses raw attention logits via Relative Contextualization (RC), a random variable measuring contextual alignment between token segments, and derive an efficient upper bound for RC. We demonstrate two applications: (i) Key-Value compression, where RC-based thresholds drive adaptive key-value eviction for substantial cache reduction with minimal quality loss; and (ii) Attribution, where RC yields higher-fidelity token-, sentence-, and chunk-level explanations than post-Softmax methods. Across question answering, summarization, and attribution benchmarks, RCStat achieves significant empirical gains, delivering state-of-the-art compression and attribution performance without any model retraining.