NDCG@10
PulseAugur coverage of NDCG@10 — every cluster mentioning NDCG@10 across labs, papers, and developer communities, ranked by signal.
8 day(s) with sentiment data
-
Component-aware feedback boosts LLM program evolution efficiency
Researchers have developed a new method called component-aware feedback to improve the efficiency of LLM-guided evolutionary search for program development. This technique logs changes made to program components and the…
-
New methods boost visual document reranking efficiency and accuracy
Researchers have developed two novel methods, RenderRank and RidgeRank, for efficient visual document reranking. RenderRank utilizes compressed visual tokens derived from document images to learn query-dependent relevan…
-
New LLM scoring method reduces order dependence in decisions
A new research paper introduces Order-Consistency Supervised Fine-Tuning (OC-SFT), a method designed to reduce order dependence in Large Language Model (LLM) scorers. These scorers, used in tasks like passage reranking …
-
New benchmark improves SNOMED CT concept recommendation from clinical text
Researchers have developed a new benchmark for recommending SNOMED CT concepts from masked clinical contexts, aiming to improve the standardization and interoperability of clinical language. The study utilized the SNOME…
-
New single-token scoring method improves LLM candidate ranking
Researchers have developed a new method called single-token expected-value scoring for ranking job candidates, particularly useful for platforms with limited interaction data. This technique casts candidate-job relevanc…
-
Research paper questions effectiveness of post-training quantization for text embedders
A new research paper published on arXiv investigates the effectiveness of post-training quantization (PTQ) techniques for text embedders. The study found that common heuristics for PTQ, such as protecting the embedding …
-
MoE models get smarter pruning, retrieval, and inference efficiency
Researchers are exploring advanced techniques for Mixture-of-Experts (MoE) language models to improve their efficiency and performance. One paper introduces HOPE (Higher-Order Pruning of Experts), a novel pruning object…
-
Polite prompts alter LLM relevance judgments, study finds
A new study published on arXiv investigates the impact of politeness in prompts on large language models used as relevance judges. Researchers found that tone significantly affects model judgments, with effects varying …
-
Tendero eCommerce launches with .NET 11 and AI layer
Tendero eCommerce, formerly CocktailDev, has launched as a real-world online shop built with .NET 11 and Angular 22. The platform incorporates an AI layer and aims to enhance the user experience. Key features include a …
-
LLM components critically shape conversational recommendation performance
A new research paper explores how different components of Large Language Models (LLMs) impact their performance and stability when used in conversational recommendation systems. The study found that proprietary LLMs sig…
-
New method constructs tree knowledge graphs for low-resource languages
Researchers have developed an end-to-end pipeline for constructing tree knowledge graphs and evaluating retrieval strategies, specifically addressing the challenge of low-resource languages like Vietnamese. The system e…
-
New OC-SFT method improves LLM scorer stability, reducing order dependence
Researchers have developed a new training method called Order-Consistent Scoring (OC-SFT) to address the issue of order dependence in Large Language Model (LLM) scorers. While existing scorers are evaluated on ranking q…
-
New method adapts dense retrieval models using agent search traces
Researchers have developed Navigation-Informed Embeddings (NIE), a novel method to adapt dense retrieval models using traces from agentic search workflows. This approach leverages query, retrieval, and stopping traces t…
-
ConceptFormer framework improves visual document retrieval using latent concepts
Researchers have introduced ConceptFormer, a novel framework designed to enhance visual document retrieval. This approach learns adaptive latent concepts to better align queries with relevant documents, even when eviden…
-
New method boosts temporal retrieval accuracy using difficulty-gated reasoning
Researchers have developed a novel method for improving temporal retrieval in information retrieval systems. This technique, called difficulty-gated fusion of reasoning views, addresses the challenge of matching queries…
-
UniMoMo framework compresses MoE recommendation models for faster inference
Researchers have developed UniMoMo, a post-training compression framework designed to accelerate large recommendation models that utilize mixture-of-experts (MoE) layers. This method groups experts based on their functi…
-
PLAID-PRF enhances dense retrieval with centroid-aware pseudo-relevance feedback
Researchers have introduced PLAID-PRF, a novel method for enhancing multi-vector dense retrieval models like ColBERT. This technique leverages centroid-like tokens within the PLAID framework to perform Pseudo-Relevance …
-
Apple TV search system uses hybrid embeddings to boost personalization
Apple's machine learning research team has developed a personalization system for Apple TV search that enhances incremental search results. The system combines text-based and ID-based embeddings, trained using contrasti…
-
New method boosts SQL schema retrieval accuracy
Researchers have developed a new method for SQL schema retrieval, focusing on identifying relevant tables and columns for natural language queries. They adapted existing text-to-SQL datasets into retrieval tasks and fou…
-
New Shard method enhances privacy in dense retrieval systems
Researchers have developed a new method called Shard to enhance privacy in dense retrieval systems, which are commonly used for semantic search and retrieval-augmented generation (RAG). Shard addresses the vulnerability…