DiscourseFlip is a graph-guided attack allocating limited poisoning budget to induce targeted opinion shifts over semantic query networks in black-box RAG.
super hub Mixed citations
Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models
Mixed citation behavior. Most common role is background (46%).
284
Pith papers citing it
Background
46% of classified citations
abstract
In this work, we introduce the Qwen3 Embedding series, a significant advancement over its predecessor, the GTE-Qwen series, in text embedding and reranking capabilities, built upon the Qwen3 foundation models. Leveraging the Qwen3 LLMs' robust capabilities in multilingual text understanding and generation, our innovative multi-stage training pipeline combines large-scale unsupervised pre-training with supervised fine-tuning on high-quality datasets. Effective model merging strategies further ensure the robustness and adaptability of the Qwen3 Embedding series. During the training process, the Qwen3 LLMs serve not only as backbone models but also play a crucial role in synthesizing high-quality, rich, and diverse training data across multiple domains and languages, thus enhancing the training pipeline. The Qwen3 Embedding series offers a spectrum of model sizes (0.6B, 4B, 8B) for both embedding and reranking tasks, addressing diverse deployment scenarios where users can optimize for either efficiency or effectiveness. Empirical evaluations demonstrate that the Qwen3 Embedding series achieves state-of-the-art results across diverse benchmarks. Notably, it excels on the multilingual evaluation benchmark MTEB for text embedding, as well as in various retrieval tasks, including code retrieval, cross-lingual retrieval and multilingual retrieval. To facilitate reproducibility and promote community-driven research and development, the Qwen3 Embedding models are publicly available under the Apache 2.0 license.
hub tools
citation-role summary
background 14
method 6
baseline 3
dataset 1
citation-polarity summary
claims ledger
- abstract In this work, we introduce the Qwen3 Embedding series, a significant advancement over its predecessor, the GTE-Qwen series, in text embedding and reranking capabilities, built upon the Qwen3 foundation models. Leveraging the Qwen3 LLMs' robust capabilities in multilingual text understanding and generation, our innovative multi-stage training pipeline combines large-scale unsupervised pre-training with supervised fine-tuning on high-quality datasets. Effective model merging strategies further ensure the robustness and adaptability of the Qwen3 Embedding series. During the training process, the
authors
co-cited works
representative citing papers
A new corpus of 108 mixed string-numeric tables shows that advanced tabular learners with basic string embeddings perform well on most real-world data, while large LLM encoders help on free-text heavy tables.
SLAM achieves 100% detection on Gemma-2 models with only 1-2 point quality cost by causally steering SAE-identified residual-stream directions for linguistic structure.
ReasonAudio benchmark reveals that state-of-the-art text-audio retrieval models struggle with reasoning tasks like negation and duration, and multimodal LLMs lose reasoning ability after contrastive fine-tuning.
FollowTable is the first large-scale benchmark for instruction-following table retrieval, paired with an Instruction Responsiveness Score, showing that existing models fail to adapt to fine-grained constraints beyond topical similarity.
MoHallBench is a new benchmark evaluating motion hallucination in VideoLLMs from co-occurrence priors, sequential inference, and similarity confusion, revealing decoupling from action recognition performance.
STEB is a new benchmark of 96 datasets in 7 languages for evaluating style text embeddings on authorship, detection, and linguistic probing tasks.
Tabular foundation models excel on tiny- to medium-sized IID data but are outperformed by traditional tree-based and deep learning models on non-IID, large, and high-dimensional datasets, based on evaluations across 11 models and 142 datasets in the new BeyondArena benchmark.
Turn-averaged SAEs reconstruct average activations over conversation turns to represent high-level turn characteristics with a fixed number of features, simplifying long-context interpretability compared to per-token SAEs.
A Gaussian information-gain metric in embedding space quantifies semantic progress in dialogues via uncertainty reduction and shows competitive agreement with human judgments on MT-Bench and UltraFeedback.
EBA clusters sampled LLM generations in representation space to estimate agreement, outperforming random selection with stable scaling and showing that central positions correlate with higher generation quality.
TAA-k finds query-adaptive retrieval cutoffs by first using knee detection to isolate a candidate window around the relevance-to-noise transition, then applying EVT goodness-of-fit tests inside that window.
CORE-Bench is a benchmark for code retrieval in agentic coding settings, built from curated tasks and SWE-bench instances, showing performance drops and gains from fine-tuning.
ActProbe is an action-space detector that uses temporal consistency error and action chunk magnitude from policy outputs, mapped via LSTM-MLP, to predict failures earlier than baselines across policies and real-robot tasks.
SEA-Embedding is a fully open text embedding pipeline for Southeast Asian languages that achieves state-of-the-art performance on the SEA-BED benchmark by analyzing data composition, training objectives, and base encoder choices.
Defines cost-aware RAG with evidence cost tiers and shows static selectors are brittle while agentic LLM-based selection is promising but model-dependent.
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.
Sakura is a multi-agent system that generates structurally complex tests from NL descriptions, achieving 50-78% higher compilability and 38-66% higher coverage overlap than baselines on 1,464 scenarios from 20 Apache Commons applications.
HEART-Bench evaluates LLM agents on psychological consistency using 11 Big-Five-grounded characters with 1,000 episodic memories each and 64 DIAMONDS-based decision scenarios, yielding 673 validated MCQs.
VeriTrip is a new benchmark using a Multimodal Retrieval Base and Verifiable Knowledge Base to evaluate evidence-grounded reasoning and factual reliability in travel planning agents over unstructured multimodal web data.
RoBatch is a two-stage framework that formulates and solves the joint Route with Batching Problem via a batch-aware proxy utility model and greedy scheduling, outperforming separate routing or batching baselines on six benchmarks.
HTEB introduces dynamic, multi-axis evaluation of text embedding robustness using LLM transformations, finding decoupled profiles across models and that scaling does not close all robustness gaps.
LLM-Wiki structures external knowledge as compilable wiki pages with links and persistent self-correction, achieving SOTA results on HotpotQA, MuSiQue, and 2WikiMultiHopQA by 2.0-8.1 F1 points over prior RAG systems.
Semantic-ID tokenizers produce collisions affecting up to 30.5% of items across four datasets, inflating Hit@10 by up to 103.36% and making prior tokenizer comparisons unreliable.
citing papers explorer
-
ReTool-Video: Recursive Tool-Using Video Agents with Meta-Augmented Tool Grounding
ReTool-Video uses a 134-tool meta-augmented library and recursive grounding to translate abstract video intents into fine-grained multimodal operations, outperforming baselines on MVBench, MLVU, and Video-MME.
-
Evolving-RL: End-to-End Optimization of Experience-Driven Self-Evolving Capability within Agents
Evolving-RL jointly optimizes experience extraction and utilization in LLM agents via RL with separate evaluation signals, delivering up to 98.7% relative gains on out-of-distribution tasks in ALFWorld and Mind2Web.
-
LatentRAG: Latent Reasoning and Retrieval for Efficient Agentic RAG
LatentRAG performs agentic RAG by generating latent tokens for thoughts and subqueries in one forward pass, matching explicit methods' accuracy on seven benchmarks while reducing latency by ~90%.
-
On the Robustness of LLM-Based Dense Retrievers: A Systematic Analysis of Generalizability and Stability
LLM-based dense retrievers generalize better when instruction-tuned but pay a specialization tax when optimized for reasoning; they resist typos and corpus poisoning better than encoder-only baselines yet remain vulnerable to semantic perturbations, with larger models and certain embedding geometry,
-
Hidden in the Multiplicative Interaction: Uncovering Fragility in Multimodal Contrastive Learning
Multimodal contrastive learning using multilinear products is fragile to single bad modalities, and a gated version improves top-1 retrieval accuracy on synthetic and real trimodal data.
-
Characterizing and Mitigating False-Positive Bug Reports in the Linux Kernel
False-positive bug reports in the Linux kernel consume effort comparable to real bugs and can be filtered by LLMs using retrieval-augmented generation at 88% F1.
-
Aspect-Aware Content-Based Recommendations for Mathematical Research Papers
The authors introduce aspect-aware datasets GoldRiM and SilverRiM for math papers and AchGNN, a heterogeneous GNN that outperforms prior methods by jointly modeling textual semantics, citations, and author lineage across aspects.
-
From Local Indices to Global Identifiers: Generative Reranking for Recommender Systems via Global Action Space
GloRank reformulates list-wise reranking as token generation over a global item identifier space, using supervised pre-training followed by reinforcement learning to maximize list-wise utility and outperforming baselines on benchmarks and industrial data.
-
Lost in Decoding? Reproducing and Stress-Testing the Look-Ahead Prior in Generative Retrieval
Reproduction confirms PAG boosts generative retrieval effectiveness, but its look-ahead planning signal collapses under intent-preserving typos and query mismatches, reverting performance to unguided decoding.
-
The Art of (Mis)alignment: How Fine-Tuning Methods Effectively Misalign and Realign LLMs in Post-Training
ORPO is most effective at misaligning LLMs while DPO excels at realigning them, though it reduces utility, revealing an asymmetry between attack and defense methods.
-
GRC: Unifying Reasoning-Driven Generation, Retrieval and Compression
GRC unifies generation, retrieval, and compression in LLMs via meta latent tokens for single-pass execution with modular flexibility.
-
Efficient Listwise Reranking with Compressed Document Representations
RRK compresses documents to multi-token embeddings for efficient listwise reranking, enabling an 8B model to achieve 3x-18x speedups over smaller models with comparable or better effectiveness.
-
Understanding the Prompt Sensitivity
LLMs disperse meaning-preserving prompts internally instead of clustering them, which produces an excessively high upper bound on output log-probability differences via Taylor expansion and Cauchy-Schwarz.
-
Comparison of Modern Multilingual Text Embedding Techniques for Hate Speech Detection Task
Supervised models using embeddings like jina and e5 reach up to 92% accuracy on multilingual hate speech detection, substantially outperforming anomaly detection, while PCA to 64 dimensions preserves most performance in the supervised case.