Jul 2026· Annual International ACM SIGIR Conference on Research and Development in Information Retrieval· 0 citations· 47 references
Computer Science
TL;DR
AD-Reranker is proposed, a novel framework that shifts reranker training from proxy imitation to answer-driven utility optimization, and reformulate the reranker as an environment-grounded agent that interacts with a downstream reader, modeled as a deterministic environment.
Abstract
Multimodal Document Question Answering (MDQA) systems commonly follow a retrieve-then-answer paradigm; however, the retrieval stage often introduces substantial noise, making an effective reranking component indispensable. Existing reranker training frameworks in MDQA rely predominantly on proxy supervision derived from human annotations or large language model (LLM) outputs, which are frequently noisy and, more critically, misaligned with downstream answer quality. To overcome this limitation, we propose AD-Reranker, a novel framework that shifts reranker training from proxy imitation to answer-driven utility optimization. Specifically, we reformulate the reranker as an environment-grounded agent that interacts with a downstream reader, modeled as a deterministic environment. We further design a composite reward function that integrates answer correctness, thereby explicitly incentivizing ranking strategies aligned with downstream task performance. To optimize the agent, we adopt Group Relative Policy Optimization (GRPO), enabling stable and effective group-wise policy learning. Empirical results demonstrate that AD-Reranker achieves superior reranking quality and an optimal accuracy-efficiency trade-off. When integrated into standard MDQA pipelines, AD-Reranker improves accuracy by 1.9%–5.0% while reducing the reader's context token consumption by 15%–52%, providing strong evidence for the effectiveness of answer-driven reranker training.
TALRanker is a novel framework that formalizes pointwise relevance scoring as an agentic Markov decision process that achieves state-of-the-art performance across standard and reasoning-intensive retrieval benchmarks, matching throughput with pointwise rerankers while outperforming parameter-heavy reasoning models.
Guided Retrieval Training (GRT) is introduced, a novel method that improves the performance of a search agent by restricting the retrieval process during RL training using ground truth information, and enhances training efficiency by achieving better QA performance with fewer training steps.
Aounon Kumar, Sudipta Paul, Vivek Kulkarni et al.· 0 citations
E-commerce search ranking models face the challenging and critical problem of balancing strict real-time latency constraints with the need for high-quality relevance predictions. In production environments, ranking models often rely primarily on product titles, which frequently omit critical attributes required to satisfy diverse query intents. While full product descriptions provide richer information, their length and verbosity make them computationally impractical for real-time ranking, particularly when using cross-encoder architectures. To address this challenge, we propose ReLSum, a reinforcement learning framework that generates concise, relevance-optimized product summaries for search ranking. ReLSum directly aligns summarization with the ranking objective by using downstream relevance scores as reward signals. The framework conditions the Large Language Model (LLM) solely on product information, while queries are used only to compute rewards during training. This design enables summaries to be generated and cached offline, ensuring no additional inference latency at serving time. Experiments on large-scale production data show substantial improvements in offline NDCG and recall. In online A/B tests, ReLSum delivers statistically significant gains in user engagement metrics such as orders per visitor and units per completed order, with particularly strong improvements for tail queries.
Evaluation on WikiSA and ExaRank shows that ranking-based few-shot prompting generally improves over zero-shot prompting and achieves competitive performance against random-shot prompting, indicating that retrieval-based demonstration selection is beneficial but not uniformly superior in all settings.
A. Laksito, Aali Alqarni, Mark Stevenson· International Conference on...· 0 citations
Efficient multimodal document question answering with explicit visual grounding, locating the precise document region that supports each answer remains an open challenge. Current approaches bifurcate into Supervised Fine-Tuning (SFT), which requires large annotated datasets and reaches optimization plateaus, and reasoning-centric Reinforcement Learning (RL), which depends on verbose intermediate traces that inflate inference token cost without clear benefit. We introduce Perception-RFT, a training framework that applies Group Relative Policy Optimization (GRPO) to multimodal document QA, bypassing intermediate reasoning tokens to directly align visual features with structured grounding outputs. To rigorously evaluate the necessity of reasoning, we construct a reasoning variant under identical reward settings. We find that reasoning-enabled models suppress their reasoning traces during training, converging to direct perception-based policies at the 4B parameter scale, reducing per-query inference token length by more than 60%, while reasoning-enabled RL underperforms perception-only training. Through a fine-grained analysis of Qwen3-VL-4B optimization dynamics, we confirm that SFT saturation and cold-start RL instability established in text-domain post-training extend to multimodal, and identify a previously uncharacterized Grounding Divergence: a selective trade-off between semantic robustness and geometric precision on two out of distribution (OOD) benchmarks (4,828 samples) under joint RL optimization. We further show that an early SFT$\rightarrow$RL transition achieves comparable precision with 65% less training data.
M. HarikrishnanP., G. Vignesh, Ganesh Parab et al.· 0 citations
E-commerce search relevance modeling faces a critical dilemma: traditional models falter with complex queries, while Large Language Models (LLMs), despite their superior reasoning, suffer from the prohibitive latency of auto-regressive Chain-of-Thought (CoT) generation, rendering them infeasible for production. Knowledge distillation offers a promising solution, yet current methods force an undesirable trade-off: sacrificing the very interpretability that makes LLMs powerful, or relying on expensive, unscalable human-annotated rationales. To address this, we propose STAR—Staged Training with Aligned Reinforcement Learning and Multi-Faceted Distillation, a progressive framework that follows a reasoning, ranking, and transfer pipeline to imbue dense models with both high performance and interpretability. First, STAR aligns a teacher LLM's reasoning with task objectives using a novel multi-granularity reward in Group Relative Policy Optimization (GRPO), leveraging only binary labels. Next, it refines the teacher's ability for calibrated scoring via token-level supervision, enabling efficient ranking through a single forward pass without any additional layers. Finally, this ''white-box'' knowledge is transferred to a compact student via multi-faceted distillation that preserves both reasoning logic and ranking behavior. Offline experiments demonstrate that our 0.6B student model rivals the performance of a strong 8B baseline, making it highly efficient and fully deployable. Real-world effectiveness is validated by significant online A/B test gains, including a +0.93% GoodRate lift and a +1.04% increase in GMV. STAR has been fully deployed to 100% of main search traffic on 1688.com.
Chenxu Wang, Jianzhi Shao, Chi Zhang et al.· Annual International ACM SIG...· 0 citations