Self-generated QA supervision for language models is fragile due to non-uniform question selection and instruction compliance during answering, with mitigations that reduce compliance from 88% to 13%.
Canonical reference
Promptagator: Few-shot dense retrieval from 8 examples
Canonical reference. 100% of citing Pith papers cite this work as background.
11
Pith papers citing it
Background
100% of classified citations
citation-role summary
background 6
citation-polarity summary
roles
background 6polarities
background 6representative citing papers
UnIte selects target-domain documents for pseudo-query generation by filtering high aleatoric uncertainty and prioritizing high epistemic uncertainty, yielding +2.45 to +3.49 nDCG@10 gains on BEIR with ~4k samples.
A survey that defines Compound AI Systems, proposes a multi-dimensional taxonomy based on component roles and orchestration strategies, reviews four foundational paradigms, and identifies key challenges for future research.
SPECTRA generates reproducible synthetic IR corpora up to 60,000 documents with controllable distractors, long-tail vocabulary, and graded relevance labels via a single-process Python prototype.
ARHN refines hard-negative training data for dense retrieval by using LLMs to convert answer-containing passages into additional positives and exclude answer-containing passages from the negative set.
Empirical comparison across 14 retrievers on the BRIGHT benchmark shows reasoning-specialized models can match strong accuracy with competitive speed while many large LLM bi-encoders add latency for small gains and confidence scores remain poorly calibrated.
RankZephyr is a new open-source LLM that closes the effectiveness gap with GPT-4 for zero-shot listwise reranking while showing robustness to input ordering and document count.
LLM-generated reference documents serve as relevance pivots for dynamic ranked-list truncation and adaptive/parallel listwise reranking, reportedly beating prior RLT methods and cutting LLM reranking cost by up to 66%.
E5 text embeddings trained with weakly-supervised contrastive pre-training on CCPairs outperform BM25 on BEIR zero-shot and achieve top results on MTEB, beating much larger models.
Multi-objective LTR combining clicks, VLM labels, and locale boosting improves relevance and local content visibility across five growth markets.
A survey of RAG paradigms, components, benchmarks, and challenges for improving LLMs on knowledge-intensive tasks.
citing papers explorer
-
Self-Study Reconsidered: The Hidden Fragility of Learning from Self-Generated QA
Self-generated QA supervision for language models is fragile due to non-uniform question selection and instruction compliance during answering, with mitigations that reduce compliance from 88% to 13%.
-
UnIte: Uncertainty-based Iterative Document Sampling for Domain Adaptation in Information Retrieval
UnIte selects target-domain documents for pseudo-query generation by filtering high aleatoric uncertainty and prioritizing high epistemic uncertainty, yielding +2.45 to +3.49 nDCG@10 gains on BEIR with ~4k samples.
-
From Standalone LLMs to Integrated Intelligence: A Survey of Compound Al Systems
A survey that defines Compound AI Systems, proposes a multi-dimensional taxonomy based on component roles and orchestration strategies, reviews four foundational paradigms, and identifies key challenges for future research.
-
SPECTRA: Synthetic IR Test Collections with Relevance Oracles and Controlled Distractor Diagnostics
SPECTRA generates reproducible synthetic IR corpora up to 60,000 documents with controllable distractors, long-tail vocabulary, and graded relevance labels via a single-process Python prototype.
-
ARHN: Answer-Centric Relabeling of Hard Negatives with Open-Source LLMs for Dense Retrieval
ARHN refines hard-negative training data for dense retrieval by using LLMs to convert answer-containing passages into additional positives and exclude answer-containing passages from the negative set.
-
Are LLM-Based Retrievers Worth Their Cost? An Empirical Study of Efficiency, Robustness, and Reasoning Overhead
Empirical comparison across 14 retrievers on the BRIGHT benchmark shows reasoning-specialized models can match strong accuracy with competitive speed while many large LLM bi-encoders add latency for small gains and confidence scores remain poorly calibrated.
-
RankZephyr: Effective and Robust Zero-Shot Listwise Reranking is a Breeze!
RankZephyr is a new open-source LLM that closes the effectiveness gap with GPT-4 for zero-shot listwise reranking while showing robustness to input ordering and document count.
-
Dynamic Ranked List Truncation for Reranking Pipelines via LLM-generated Reference-Documents
LLM-generated reference documents serve as relevance pivots for dynamic ranked-list truncation and adaptive/parallel listwise reranking, reportedly beating prior RLT methods and cutting LLM reranking cost by up to 66%.
-
Text Embeddings by Weakly-Supervised Contrastive Pre-training
E5 text embeddings trained with weakly-supervised contrastive pre-training on CCPairs outperform BM25 on BEIR zero-shot and achieve top results on MTEB, beating much larger models.
-
Localization Boosting for Growth Markets: Mitigating Cross-Locale Behavioral Bias in Learning-to-Rank
Multi-objective LTR combining clicks, VLM labels, and locale boosting improves relevance and local content visibility across five growth markets.
-
Retrieval-Augmented Generation for Large Language Models: A Survey
A survey of RAG paradigms, components, benchmarks, and challenges for improving LLMs on knowledge-intensive tasks.