cs.IR
PulseAugur coverage of cs.IR — every cluster mentioning cs.IR across labs, papers, and developer communities, ranked by signal.
4 day(s) with sentiment data
-
New RAG framework tackles OCR noise in historical archives
Researchers have developed a novel framework to improve information retrieval from historical legislative archives, specifically addressing challenges posed by degraded documents and OCR errors. Their system integrates …
-
DyHist framework enhances temporal knowledge graph extrapolation
Researchers have developed DyHist, a new framework designed to improve the prediction of future facts from historical data in temporal knowledge graphs (TKGs). DyHist addresses limitations in existing methods by explici…
-
LLM investor simulations lack individual-level fidelity, study finds
A new study published on arXiv investigates the ability of large language models (LLMs) to accurately simulate individual investor behavior over time. In a paper-trading simulation involving 80 participants, researchers…
-
AI agent memory failures predicted by geometric analysis, but with limited success
Researchers explored whether geometric relationships within conversation data could predict retrieval-bound memory failures in AI agents before generation. They embedded conversation turns, memory texts, and queries int…
-
Biomedical RAG performance hinges on corpus scale and document representation, study finds
A new study evaluates the impact of corpus scale, document representation, and reranking on the performance and efficiency of biomedical Retrieval-Augmented Generation (RAG) systems. The research found that while RAG ge…
-
New research proposes Confluent Evidence Acquisition for AI agents
A new research paper introduces Confluent Evidence Acquisition (CEA), a method designed to improve how language-model agents gather information during complex tasks. Current agents can be overly influenced by the timing…
-
New framework improves survey data harmonization with concept-aware retrieval
Researchers have developed a new conditional information retrieval framework to improve the harmonization of longitudinal survey data. This framework first extracts underlying sociological concepts from survey questions…
-
FLMNet advances CTR prediction with novel self-distillation technique
Researchers have developed FLMNet, a novel approach to click-through rate (CTR) prediction that improves accuracy by learning from averaged predictions without relying on random sampling. This method, inspired by self-d…
-
New MOSS framework enhances recommender system diversity
Researchers have developed a new framework called MOSS to improve the diversity of recommender systems. Current systems often prioritize accuracy, leading to a narrow range of recommendations. MOSS addresses this by dec…
-
New benchmark shows SQL execution boosts RAG accuracy for quantitative materials science research
A new benchmark for structured retrieval-augmented generation (RAG) in perovskite photovoltaics has been developed, focusing on quantitative answers derived from experimental data rather than qualitative text overlap. T…
-
New methods and frameworks advance MLLM unlearning, improving evaluation and knowledge suppression
Researchers are developing new methods to improve the unlearning process for multimodal large language models (MLLMs), aiming to ensure that models not only stop outputting specific information but also genuinely lose t…
-
New 'Quantize by Drift' method optimizes text embedders without labels
Researchers have developed a novel method called "Quantize by Drift" for optimizing text embedders without requiring relevance labels. This technique measures the impact of quantization on embedding representations, usi…
-
New research explores optimal data compression using quantum retrieval
A new research paper explores the theoretical limits of data compression using quantum retrieval methods. The study focuses on compressing a binary string with a limited number of set bits into a shorter string, allowin…
-
New semantic indexing method 5W1H+Which separates content from ontology binding
Researchers have introduced a novel semantic indexing method called 5W1H+Which, designed to improve the queryability of raw data by separating content extraction from ontology binding. This approach organizes source-gro…
-
New PCGNet framework unifies fashion compatibility and user preferences
Researchers have developed PCGNet, a novel graph learning framework designed to improve fashion matching recommendations. This multi-objective system unifies product compatibility and individual user preferences by extr…
-
New ensemble method boosts vacation rental recommendations with GNNs · 2 sources tracked
Researchers have developed a new ensemble approach for generating alternative vacation rental property recommendations, significantly improving recall rates. This method combines item-based collaborative filtering with …
-
New research reveals Sublinear Power Law in vector search scalability
A new paper published on arXiv introduces the "Sublinear Power Law" to describe the scalability of graph-based vector search. Researchers found that search cost grows as N^c (where c is less than 1) when dataset size (N…
-
New Adaptive Doubly Robust Method Enhances Off-Policy Evaluation for Ranking Policies
Researchers have introduced Adaptive Doubly Robust (ADR), a novel method for off-policy evaluation (OPE) of ranking policies. ADR aims to reduce the variance and bias inherent in existing OPE techniques like Inverse Pro…
-
PailitaoGR method enhances generative image retrieval by focusing on targets and auxiliary evidence · 2 sources tracked
Researchers have introduced PailitaoGR, a novel method for generative image retrieval that enhances the model's ability to focus on search targets and selectively utilize auxiliary visual information. This approach aims…
-
W-RAG framework improves enterprise document generation with source-aware retrieval
Researchers have introduced W-RAG, a novel framework designed to enhance enterprise document generation by improving retrieval-augmented generation (RAG) pipelines. Unlike standard RAG that uses a single similarity func…