Research
Papers and datasets worth knowing, ranked by significance and community attention.
OneSearch-VL: Unified Multimodal Deep Research Agent for Image and Video
We introduce OneSearch-VL, a unified agent centered on the Visually Grounded Evidence Graph (VGEG), which encodes these dependencies as a shared task-level reference for data construction, process supervision, and operation-level evaluation.
RATED: Reliability-Aware AI Agent Tool Retrieval via Execution Dynamics
To address this gap, we propose RATED (Reliability-Aware Tool Retrieval via Execution Dynamics), a retrieval framework that combines robust semantic matching with dynamic reliability adaptation.
Can Language Models Play the Wikipedia Game?
Language models increasingly support or automate information retrieval from document collections too large to index.
Knowing Whether, Not Just Where: Anchor-Probing Distillation for Open-Set Remote Sensing Visual Grounding
Open-set visual grounding in remote sensing requires models to determine not only where a queried object is located, but also whether the image provides sufficient evidence for the query at all.
PlacePulse: Learning Transferable Place Representations from Structural Geographic Data
We introduce PlacePulse, a structural geospatial foundation model that compresses 572 attributes, expanded to 2,288 inputs through H3 neighborhood aggregates, into 256-dimensional unit-norm embeddings for 1,055,537 H3 resolution-7 cells in the United States.
Host-Aware Web Crawling for LLM Pretraining Data
Web crawling is a major source of large language model (LLM) pretraining data, yet data-selection pipelines discard a large share of crawled pages.
Marformer: A Transformer for Predicting Missing Data Distributions
We present the Marformer, a Transformer trained to directly predict conditional marginals given any set of observed values.
Incremental Retrieval-Augmented Generation for Human-in-the-Loop Decision Support
We introduce iRAG, a human-in-the-loop decision-support system combining Retrieval-Augmented Generation (RAG) with Hybrid Decision-Making.
Verifiable Retrieval for Legal Reasoning through Structure-driven Knowledge Graph
Using structure-driven parsing to create a Knowledge Graph (KG) for the source document navigation and retrieval, we compare several KG-informed chunking and retrieval strategies and evaluate their effect on the accuracy of retrieved citations.
Multi-Stream Attention Residuals
Inspired by the success of multi-head attention along the sequence axis, we introduce Multi-Stream Attention Residuals (MSAR), an extension of AttnRes that generalizes this mechanism to multi-head depth attention, yielding substantial performance gains with minimal additional overhead.
Tracing and Taming Hallucinated Terms via Subsequence-Association Filtering and Breadth-Aware Selection for LLM Query Expansion
We propose Subsequence-Association Filtered, breadth-aware query expansion (SAFE-QE), a training-free, retriever agnostic method that draws on the subsequence-association account of hallucination.
Representation Alignment from a Few Paired Data via Distilled Optimal Transport
Aligning two independently pretrained unimodal encoders typically demands millions of paired examples.
PRIME: Pre-Retrieval Index Modeling Enhancement for Structured Document Retrieval-Real
We propose Pre-Retrieval Index Modeling & Enhancement (PRIME), an index-side optimization framework built on two complementary tracks: an LLM-driven Semantic Restructuring (SR) track that transforms long documents into structured, information-dense sections, and a deterministic Metadata Synthesis (MS) and Hierarchical Retrieval (HR) track that extracts semantic anchors and heading structure for coarse-to-fine section retrieval.
Locate, Judge, Read: Role-Specialized Tool Agents over a Knowledge-Point Index for Open-Domain Multimodal Document QA
We present KnowForge, a pipeline separated by job: a hybrid-retrieval knowledge-point index; a Locator agent that can only search; a code-scheduled bridge judge that forces the missing second hop; a typed, size-bounded evidence pool that merges every search's ranking; a Deep Reader agent that can only read text, images and tables; and a synthesizer with a format checker.
Personalization Matters: Long-Horizon Conversation Agent with User-Centric Information in Online Shopping Interactions
We propose a multi-agent, multimodal Retrieval-Augmented Generation (RAG) framework that decomposes dialogue state tracking, recommendation retrieval, preference-aware reasoning, and response generation, while integrating product metadata, product reviews, image-derived descriptions, and user historical reviews.
RIT-RAG: Navigating Document Corpora with Retrieval-Induced Trees
Retrieval-augmented generation (RAG) grounds language models in external corpora.
RAG-Stress: Probing the Limits of Evidence Reliance in Retrieval-Augmented Generation
We introduce RAG-Stress, a controlled diagnostic protocol for examining the limits of evidence reliance in retrieval-augmented generation.
Evolving Creatives with Clicks: CTR-Driven Advertising image Editing via Attribute Disentanglement and Preference Alignment
In this work, we propose Evolving Creatives with Clicks (ECC), which reformulates creative optimization as fine-grained editing rather than generation from scratch.
CLUE: A Contrastive Learning Framework for Unsupervised Event Type Induction and Annotation
To alleviate the need for manual event ontology creation, we propose the CLUE framework for event type induction, naming, and annotation.
Compile the Table: Query-Calibrated Operator Compression for Tabular In-Context Learning
We propose QCOC (Query-Calibrated Operator Compression), which exploits the exchangeability and repeated use of in-context examples by compiling their full KV cache once into compact memory shared across subsequent queries.
GeopoliticalQA: Benchmarking Retrieval and Generation over Intelligence Reports
We introduce a benchmark for grounded question answering over 415 open-source geopolitical and strategic intelligence reports, comprising 5,525 document-grounded question-answer pairs spanning factual, inferential, comparative, and risk-oriented questions.
NativeScope: Relation-Localized Retrieval over Native Topology with a Correct Anchor
We propose NativeScope, a scope-then-rank method for queries with a known anchor and relation.
Chaos in the Text: Revealing the Modality Preference in Mixed-Modality Retrievers
Dense retrievers have made significant progress on text and image corpora, but whether these capabilities extend reliably to mixed corpora containing text, image, and fused text-image documents remains unclear.
Softmax Attention on Gaussian Mixtures: Linear When It Can, Selective When It Must
This work studies the infinite-prompt limit of softmax attention on Gaussian mixtures, which retain the tractability of Gaussian data while introducing latent structure, multimodality, and nonlinear dependencies.
Benchmarking Answer-Grounding Consistency in Remote Sensing Visual Question Answering
In grounded remote sensing visual question answering, a model produces a textual answer and identifies relevant objects in the image as grounding evidence.
Project Greenhouse: Progress Toward Fully Open and Sovereign Agentic Search
Project Greenhouse represents our exploration of a simple thesis: We believe that it is possible to build fully open and sovereign models for agentic search with only modest computational resources.
Beyond the Final Embedding: Retrieving Memories from Token-Level Evidence
Memory-augmented assistants retrieve evidence from stored conversations and documents to answer queries.
EntroPrefill: Renyi-Guided Context Pruning with Conditional Stability Guarantees for Retrieval-Augmented Generation
Mid-prefill pruning can reduce the sequence processed by deeper transformer layers, but attention concentration alone does not certify that discarded context is dispensable.
Beyond a Single Crop: Contextualized Embeddings for Player Re-Identification
Here, we propose ContextReID, a contextualized embedding framework that represents a player by jointly considering other players in the same frame and observations from different frames of the broadcast sequence.
From Retrieval to Reconstruction: Constructing Evolvable Cognitive Memory for Long-Term Dialogue
Large Language Models (LLMs) serving as long-term dialogue agents require memory systems that support reliable reasoning over extended interactions.
Task-Sufficient Contraction: Source Selection for Machine Information Interfaces
This paper studies when one such reduction preserves the complete downstream problem family, a property termed Task-Sufficient Contraction.
Autoregressive Retriever: Improving Query Understanding from Item Feedback for Universal Multimodal Retrieval
We introduce the AutoRegressive Retriever (ARR), a multimodal retrieval model that learns both to select informative items and to use their content to refine subsequent retrieval.
Beyond Resolution: Object-to-Image Ratio Mismatch in Instance Retrieval
We show that the dominant cause is usually not resolution loss but object-to-image (O2I) ratio mismatch: the object occupies different fractions of the two images.
RECAST: Learning to Compute the Right Context through Adaptive Evidence Routing
In this work, we introduce RECAST (Routing Evidence through Computation, Access, and Synthesized Tools), a learned framework that formulates evidence construction as a sequential decision process over heterogeneous retrieval and computation operations, allowing evidence to be actively derived rather than merely retrieved.
CliQ-RRG: Clinical-Knowledge Guided Disease-aware Visual-Textual Alignment for QA-Style Radiology Report Generation
We present Clinical-Knowledge Guided Disease-aware Visual-Textual Alignment for QA-Style Radiology Report Generation (CliQ-RRG), a unified two-stage framework for interpretable and clinically grounded reporting.
Embedding-Bias in Conditional Independence Testing
To test conditional independence of $X$ and $Y$ given a text or an image $Z$, one conditions on an embedding $ψ(Z)$ in place of $Z$.
Beyond Sequences: Distilling Structured Decision Memory for LLM Recommendation
To bridge this gap, we propose MARI (Memory-Augmented Recommendation with Interpretability), which grounds predictions in explicit, structured decision evidence.
Benchmarking Embedding Geometries and Clustering Algorithms for Unsupervised Patent Claim Classification
This paper asks if an unsupervised representation trained on patent text can recover relevant connections that code-restricted search cannot, and if hyperbolic space helps.
Beyond Semantic IDs: Finite-Beam Reachability and Continuous-Query Generation in Generative Recommendation
Generative recommendation (GR) of social media platforms commonly retrieves items by autoregressively decoding discrete Semantic IDs (SIDs) with finite-width beam search.
SEEDS: Preview-Guided Evidence Admission for Search Agents
To enable such selective control, we introduce SEEDS (Selective Evidence Entry During Search), a preview-guided evidence admission framework for search agents.
Region-Aware CLS Token Augmentation for Fine-Grained Image Retrieval
Image retrieval methods often rely on a single global semantic descriptor extracted from an image, e.g., the [CLS] token in vision transformers.
From Chunks to Functional Evidence: Function-Aware Retrieval for EDA Documentation QA
Retrieval-Augmented Generation (RAG) is widely used to ground answers in documents.
Efficient Multi-Granularity Knowledge Transfer for Radiology Report Generation
Radiology report generation can automatically generate clinical descriptions from X-ray images, thereby significantly improving the efficiency of radiologists.
SteerCast: Retrieval-Based Latent Steering for Decoder-Only Time Series Forecasting
We propose SteerCast, a retrieval-based latent steering method that improves decoder-only forecaster at inference time, without updating its parameters.
Multimodal Graph Retrieval-Augmented Sequential Recommendation via Collaborative Filtering Paths
To address these challenges, we propose MGRASRec, a multimodal graph retrieval-augmented framework for sequential recommendation.
Semantic Views: Certified Reuse of LLM Labels Across Semantic Queries
We present SemViews, a certified semantic filter that orders tuples by signals that cost no oracle calls (the labels that earlier queries left behind as semantic views, a local embedding, and a helper model's score), refines that order with labels it must pay for anyway, and certifies its answer with fresh stratified oracle samples.
Training with Missed Targets in Generative Recommendation: Separating Supervision from Probability Competition
Generative recommenders return a limited candidate set and may omit observed targets before reranking.
KGATE : a Knowledge Graph Embedding Training Environment
Knowledge graph embedding (KGE) models encode the entities and relations of a knowledge graph into a low-dimensional latent space, enabling tasks such as classification or link prediction.