Senior MLE at Meta. Trying to keep up with the Information Retrieval domain!
Blog: https://blog.reachsumit.com/
Newsletter: https://recsys.substack.com/
Sumit
Loading...
Do Generative Recommenders Deepen the Information Cocoon? A Closed-Loop Simulation with LLM-powered User Simulators
Introduces a framework with LLM-powered user agents to study information cocoons in generative recommenders.
📝 arxiv.org/abs/2606.17707
👨🏽💻 github.com/Dregen-Yor/R...
Beyond Parallel Sampling: Diverse Query Initialization for Agentic Search
Diversifies the first-turn query across parallel search rollouts to avoid redundant retrieval, improving multi-hop QA at matched compute.
📝 arxiv.org/abs/2606.17209
👨🏽💻 github.com/cxcscmu/dive...
On the Memorization Behavior of LLMs in Generative Recommendation: Observations, Implications, and Training Strategies
Snap shows that LLM recommenders rely heavily on one-hop memorization and teaches them richer relations.
📝 arxiv.org/abs/2606.17276
👨🏽💻 github.com/snap-researc...
RSRank: Learning Relevance from Representational Shifts
Adobe introduces a lightweight reranking framework that learns relevance from the representational shift a document induces in a query's internal state, filtering irrelevant content at a natural zero threshold.
📝 arxiv.org/abs/2606.17468
Temporal Preference Optimization for Unsupervised Retrieval
Microsoft presents a preference-based training method that injects temporal awareness into unsupervised dense retrievers, helping them favor temporally aligned documents.
📝 arxiv.org/abs/2606.17664
👨🏽💻 github.com/agwaBom/TPOUR
Understanding and Debugging Failures in N-Gram-Based Generative Retrieval
Presents a taxonomy of generative retrieval failure modes & introduces a web-based tool to analyze generated n-grams and their contribution to ranking.
📝 arxiv.org/abs/2606.17721
👨🏽💻 github.com/adrianmbrach...
Non-negative Elastic Net Decoding for Information Retrieval
Introduces a retrieval method that selects documents by jointly reconstructing the query embedding as a sparse non-negative linear combination.
📝 arxiv.org/abs/2606.17910
Beyond Monolingual Deep Research: Evaluating Agents and Retrievers with Cross-Lingual BrowseComp-Plus
Introduces a benchmark that varies the language of supporting evidence while keeping English questions and answers.
📝 arxiv.org/abs/2606.15345
DEEPRUBRIC: Evidence-Tree Rubric Supervision for Efficient Reinforcement Learning of Deep Research Agents
Builds an evidence tree to jointly derive training queries and rubrics.
📝 arxiv.org/abs/2606.17029
👨🏽💻 zminghang.github.io/DeepRubric-C...
Recommender systems alleviate information overload, yet repeated feedback between recommendations and user interactions can reinforce existing preferences and narrow users' exposure, forming informati...
arxiv.org
Test-time scaling for agentic search typically increases depth (i.e., more turns and tokens per trajectory) or breadth (i.e., more parallel rollouts). Here we focus on breadth scaling, showing that st...
Generative recommendation (GR) has emerged as a promising direction for recommender systems. Recently, large language models (LLMs) have been increasingly adopted for GR, as their rich pretrained know...
arxiv.org
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 o...
Unsupervised dense retrievers offer scalability by learning semantic similarity from unlabeled documents via contrastive learning, but they struggle to capture the temporal relevance, retrieving seman...
arxiv.org
Generative Retrieval (GR) is an emerging Information Retrieval (IR) paradigm that is motivated by increasingly capable language models. In GR, a model directly generates identifiers for relevant docum...
Dense retrieval has become the dominant paradigm in information retrieval, in which each document is scored against a query by the inner product of their vector embeddings, and the top-$k$ documents b...
Deep research agents synthesize long-form reports by searching and reasoning over retrieved evidence. Reinforcement learning with rubric-based rewards improves these agents by optimizing them against ...
arxiv.org
Querit-Reranker: Training Compact Multilingual Rerankers via Efficient Label-Free Distribution Adaptation
Presents a multilingual cross-encoder reranker family trained via a data-centric pipeline for label-efficient adaptation.
📝 arxiv.org/abs/2606.19037
🤗 huggingface.co/Querit/Querit
Deep research agents are increasingly evaluated on their ability to search for evidence, reason over retrieved sources, and produce grounded answers. Existing browsing benchmarks, however, largely ass...
Deployable multilingual rerankers must generalize across languages, domains, and target ranking tasks while remaining efficient enough for second-stage reranking. However, adapting them to new target ...
