Bibliographic Explorer
PulseAugur coverage of Bibliographic Explorer — every cluster mentioning Bibliographic Explorer across labs, papers, and developer communities, ranked by signal.
17 day(s) with sentiment data
-
AI search retrieval security evaluated in new arXiv paper
A new research paper published on arXiv explores the security implications of using manifests to assist retrieval in AI search systems. The study identifies potential threat paths and evaluates the security and utility …
-
New LESSRec model enhances semantic ID-based recommendations
Researchers have introduced LESSRec, a novel approach to semantic ID-based recommendation systems. This method models historical token sequences at different semantic levels separately, focusing attention on the most re…
-
ReSight system enhances embodied agents' memory and task processing
Researchers have introduced ReSight, a novel self-improving memory system designed for embodied agents. This system allows robots to process and access historical observations more effectively as task demands evolve. Re…
-
GPT-5.5 reference generation improves with higher reasoning effort
A new study published on arXiv investigated the impact of increased reasoning effort on the reference generation capabilities of GPT-5.5. Researchers found that when GPT-5.5 operated at a higher reasoning effort, it imp…
-
Conceptizer framework trains LLMs to predict meaning, not just next token
A new training framework called Conceptizer has been developed to improve language models by rewarding the prediction of meaning, not just the next token. This approach utilizes a fixed vocabulary of approximately 9,000…
-
New framework LongSearchEval assesses LLM web search capabilities
Researchers have introduced LongSearchEval, a new framework designed to evaluate the long-horizon web search capabilities of large language models (LLMs). This framework utilizes a trajectory-aware agent protocol to col…
-
New agent PACE aids adaptive conjecture exploration in applied mathematics
Researchers have developed PACE, a novel agent designed for adaptive conjecture exploration in applied mathematics. PACE operates through a structured research loop, enabling it to select mathematical objectives and dev…
-
New blind random search algorithm tackles noisy optimization problems
Researchers have developed an enhanced blind random search algorithm designed to handle optimization problems with noisy measurements. This new method combines a vanishing acceptance threshold with a running average of …
-
New AI framework uses ontology-guided reasoning for explainable medical diagnosis
Researchers have developed a new framework for differential diagnosis that leverages multi-agent reasoning guided by medical ontologies. This approach aims to improve the explainability and accuracy of AI in clinical se…
-
New framework unifies input configuration for deep CTR prediction models
Researchers have introduced OptIDC, a novel framework designed to unify and optimize the configuration of input dimensions for deep click-through rate (CTR) prediction models. This framework addresses the limitations of…
-
New research proposes theoretically sound item similarity measures for recommender systems
A new paper proposes a principled approach to designing item similarity measures for recommender systems, moving beyond heuristic methods. The research grounds these measures in probabilistic modeling and parameter esti…
-
New self-supervised sonar imaging technique suppresses noise and blur
Researchers have developed a novel self-supervised framework for enhancing multibeam imaging sonar data. This method jointly addresses speckle noise and system-induced blur without requiring clean ground truth images. B…
-
New Survey Maps Long-Document Visual Question Answering Landscape
A new survey paper published on arXiv provides a task-centric breakdown of long-document visual question answering (VQA). It categorizes the field into document parsing, evidence organization and retrieval, generator or…
-
New method optimizes web agents' preference learning for long-term user value
Researchers have introduced Active Preference Acquisition (APA), a novel method for web agents to learn user preferences over time. Unlike passive methods that rely on revealed preferences or current-task value of infor…
-
New arXiv papers explore LLM co-evolution and classifier circuits
Two new arXiv papers explore novel ways to evaluate and improve large language models. "Environment Duels" introduces a benchmark where LLMs create and solve interactive environments, revealing that strong solvers are n…
-
Journal Basket tool aids business students in literature searches
A new browser-based tool called Journal Basket has been developed to assist business students in conducting literature searches and ensuring methodological transparency. This tool allows users to create custom journal c…
-
New research details error analysis for LLM agents
A new paper introduces a methodology for analyzing errors in Large Language Model (LLM) agents that perform multi-step tasks. The research categorizes these errors into system-level and task-level vulnerabilities, provi…
-
New encoding method optimizes data representation for machine learning features
Researchers have developed a new method for encoding data that allows for efficient evaluation with a given family of functions, inspired by the concept of features in machine learning. The proposed encoder produces a r…
-
New arXiv Paper Distinguishes Agent Harness Optimization Gains
A new paper from arXiv explores the distinction between elicitation and expansion in agent harness optimization for language models. The research analyzes whether optimized harnesses improve performance on tasks already…
-
New research paper explores state recovery via prediction on arXiv
A new research paper published on arXiv explores the concept of "ontic state recovery through prediction." The paper proposes that a system's state can be understood as its complete response profile over a given query u…