Identifies the generative-discriminative gap in LLM hard negative synthesis for retrieval and proposes CausalNeg using CoT counterfactual perturbation plus query-view entropy maximization to generate more effective negatives.
Title resolution pending
17 Pith papers cite this work. Polarity classification is still indexing.
17
Pith papers citing it
citation-role summary
background 2
citation-polarity summary
roles
background 2polarities
background 2representative citing papers
Introduces BacktestBench benchmark with 18k QA pairs across four backtesting tasks and evaluates 23 LLMs via the AutoBacktest multi-agent system.
EXG is an experience graph framework for self-evolving LLM agents that supports online real-time growth and offline reuse to enhance solution quality and efficiency on code generation and reasoning benchmarks.
SimDiff uses similarity and difference metrics to prune LLM layers more effectively than cosine similarity alone, retaining over 91% performance at 25% pruning on LLaMA2-7B.
A single query-specific poisoned document, built by extracting and iteratively refining an adversarial chain-of-thought, can substantially degrade reasoning accuracy in retrieval-augmented LLM systems.
Chain-of-Thought reasoning in LLMs is often unfaithful, with models relying on it variably by task and less so as models scale larger.
PlanRAG models natural language exploratory reasoning problems as logical query trees, optimizes them via dynamic programming with a multi-dimensional cost model, and executes iterative retrieval-generation over the trees to outperform prior RAG methods on a new dataset.
NeocorRAG uses Evidence Chains to achieve SOTA retrieval quality in RAG on HotpotQA, 2WikiMultiHopQA, MuSiQue, and NQ for 3B and 70B models while using under 20% of the tokens of comparable methods.
A human-centered design workshop with journalism practitioners yields an evaluation cookbook and design requirements for contextualized, value-aligned generative AI benchmarks.
Ishigaki-IDS is a verifier-aware LLM for generating validator-passing IDS files in BIM, reaching IDSAuditPass scores of 0.651-0.753 on a 166-case benchmark and cutting practitioner work time by 54.7%.
Representation Curriculum stages feature introduction during training to prioritize content merit signals over exposure-dependent ones, reducing shortcut learning and improving cold-start robustness in ranking.
Reproducibility study shows position and context size effects in RAG depend on topic sampling and retrieval quality, proposes calibration for stable trends, and releases code after finding discrepancies with prior industry work.
A neural sparse retrieval system with granular subword tokenization (max 3 chars) achieves 91.4% recall@10 on a 6M music document corpus versus 57.7% for trigrams, with improved HCI exploration efficiency and zero added query latency.
Passages made from high-convergence sentences improve LLM performance on inferential questions compared to cosine similarity selection.
Full-horizon planning with on-demand replanning achieves accuracy parity with single-step planning in tool-calling agents for knowledge base and multi-hop question answering while consuming 2-3 times fewer tokens.
CroSearch-R1 applies search-augmented RL with cross-lingual integration and multilingual rollouts to improve RAG effectiveness on multilingual collections.
citing papers explorer
-
When Hard Negatives Hurt: Bridging the Generative-Discriminative Gap in Hard Negative Synthesis for Retrieval
Identifies the generative-discriminative gap in LLM hard negative synthesis for retrieval and proposes CausalNeg using CoT counterfactual perturbation plus query-view entropy maximization to generate more effective negatives.
-
BacktestBench: Benchmarking Large Language Models for Automated Quantitative Strategy Backtesting
Introduces BacktestBench benchmark with 18k QA pairs across four backtesting tasks and evaluates 23 LLMs via the AutoBacktest multi-agent system.
-
EXG: Self-Evolving Agents with Experience Graphs
EXG is an experience graph framework for self-evolving LLM agents that supports online real-time growth and offline reuse to enhance solution quality and efficiency on code generation and reasoning benchmarks.
-
SimDiff: Depth Pruning via Similarity and Difference
SimDiff uses similarity and difference metrics to prune LLM layers more effectively than cosine similarity alone, retaining over 91% performance at 25% pruning on LLaMA2-7B.
-
AdversarialCoT: Single-Document Retrieval Poisoning for LLM Reasoning
A single query-specific poisoned document, built by extracting and iteratively refining an adversarial chain-of-thought, can substantially degrade reasoning accuracy in retrieval-augmented LLM systems.
-
Measuring Faithfulness in Chain-of-Thought Reasoning
Chain-of-Thought reasoning in LLMs is often unfaithful, with models relying on it variably by task and less so as models scale larger.
-
When RAG Meets Query Planning: Logical Query Trees for Resolving Exploratory Reasoning Problems
PlanRAG models natural language exploratory reasoning problems as logical query trees, optimizes them via dynamic programming with a multi-dimensional cost model, and executes iterative retrieval-generation over the trees to outperform prior RAG methods on a new dataset.
-
NeocorRAG: Less Irrelevant Information, More Explicit Evidence, and More Effective Recall via Evidence Chains
NeocorRAG uses Evidence Chains to achieve SOTA retrieval quality in RAG on HotpotQA, 2WikiMultiHopQA, MuSiQue, and NQ for 3B and 70B models while using under 20% of the tokens of comparable methods.
-
Towards Real-World Validity in Generative AI Benchmarks: Understanding and Designing Domain-Centered Evaluations for Journalism Practitioners
A human-centered design workshop with journalism practitioners yields an evaluation cookbook and design requirements for contextualized, value-aligned generative AI benchmarks.
-
Ishigaki-IDS: An Open-Weight Verifier-Aware Model for Information Delivery Specification Drafting in Building Information Modeling
Ishigaki-IDS is a verifier-aware LLM for generating validator-passing IDS files in BIM, reaching IDSAuditPass scores of 0.651-0.753 on a 166-case benchmark and cutting practitioner work time by 54.7%.
-
Representation Curriculum: Stagewise Training for Robust Ranking and Allocation
Representation Curriculum stages feature introduction during training to prioritize content merit signals over exposure-dependent ones, reducing shortcut learning and improving cold-start robustness in ranking.
-
Lost in the Evidence? Reproducing Document Position and Context Size Effects in RAG
Reproducibility study shows position and context size effects in RAG depend on topic sampling and retrieval quality, proposes calibration for stable trends, and releases code after finding discrepancies with prior industry work.
-
Surface-Form Neural Sparse Retrieval: Robust Fuzzy Matching for Industrial Music Search
A neural sparse retrieval system with granular subword tokenization (max 3 chars) achieves 91.4% recall@10 on a 6M music document corpus versus 57.7% for trigrams, with improved HCI exploration efficiency and zero added query latency.
-
Context Convergence Improves Answering Inferential Questions
Passages made from high-convergence sentences improve LLM performance on inferential questions compared to cosine similarity selection.
-
Do Agents Need to Plan Step-by-Step? Rethinking Planning Horizon in Data-Centric Tool Calling
Full-horizon planning with on-demand replanning achieves accuracy parity with single-step planning in tool-calling agents for knowledge base and multi-hop question answering while consuming 2-3 times fewer tokens.
-
CroSearch-R1: Better Leveraging Cross-lingual Knowledge for Retrieval-Augmented Generation
CroSearch-R1 applies search-augmented RL with cross-lingual integration and multilingual rollouts to improve RAG effectiveness on multilingual collections.
- Skill-RAG: Failure-State-Aware Retrieval Augmentation via Hidden-State Probing and Skill Routing