Knowledge Packs deliver knowledge via pre-computed KV caches with exact equivalence under causal masking, achieving zero divergences on tested questions and enabling value-based steering without training.
hub Canonical reference
C-Pack: Packed Resources For General Chinese Embeddings
Canonical reference. 75% of citing Pith papers cite this work as background.
47
Pith papers citing it
Background
75% of classified citations
abstract
We introduce C-Pack, a package of resources that significantly advance the field of general Chinese embeddings. C-Pack includes three critical resources. 1) C-MTEB is a comprehensive benchmark for Chinese text embeddings covering 6 tasks and 35 datasets. 2) C-MTP is a massive text embedding dataset curated from labeled and unlabeled Chinese corpora for training embedding models. 3) C-TEM is a family of embedding models covering multiple sizes. Our models outperform all prior Chinese text embeddings on C-MTEB by up to +10% upon the time of the release. We also integrate and optimize the entire suite of training methods for C-TEM. Along with our resources on general Chinese embedding, we release our data and models for English text embeddings. The English models achieve state-of-the-art performance on MTEB benchmark; meanwhile, our released English data is 2 times larger than the Chinese data. All these resources are made publicly available at https://github.com/FlagOpen/FlagEmbedding.
hub tools
citation-role summary
background 6
baseline 1
method 1
citation-polarity summary
representative citing papers
IdioLink introduces a benchmark dataset and evaluation showing that strong embedding models struggle to retrieve equivalent meanings across idiomatic and literal forms, relying on shallow cues instead.
Prism-Reranker models output relevance, contribution statements, and evidence passages to support agentic retrieval beyond scalar scoring.
HaS accelerates RAG retrieval via homology-aware speculative retrieval and homologous query re-identification validation, cutting latency 24-37% with 1-2% accuracy drop on tested datasets.
METRO induces both short-term actions and long-term planning from expert transcripts into a Strategy Forest, outperforming prior methods by 9-10% on two non-collaborative dialogue benchmarks.
DRBENCHER generates multi-hop questions across biochemistry, finance, geophysics, security, and history that test interleaved browsing and computation, where the strongest models reach only 20% accuracy and human validation finds 76% validity.
PERMA is a new benchmark using temporally ordered events, text variability, and linguistic alignment to evaluate LLM memory agents on persona consistency beyond simple retrieval.
CHIMERA is the first large-scale mined KB of concept recombinations from scientific literature, created via a new IE task and LLM extraction, with demonstrated uses in pattern analysis and hypothesis generation.
VisRAG achieves 20-40% better end-to-end performance than text-based RAG by directly embedding and retrieving document images with VLMs.
MultiHop-RAG is a new benchmark dataset demonstrating that existing retrieval-augmented generation systems perform poorly on multi-hop queries requiring retrieval and reasoning over multiple evidence pieces.
Embedding model performance on MTEB tasks correlates strongly with nearest-neighbor overlap and ICA magnitude differences in their embedding spaces.
An extended annotation scheme with new categories and attributes plus a Gemma-300M-based multi-head classifier achieves 81.6% macro F1 on personal fact classification, outperforming few-shot LLM baselines by nearly 9 points with lower compute.
SkillRAE organizes skills into a graph and compiles compact, grounded contexts for LLM agents, yielding 11.7% gains on SkillsBench over prior RAE methods.
Attention-based models can retrieve evidence intrinsically by using decoder attention to score and reuse their own pre-encoded chunks, outperforming separate retrieval pipelines on QA benchmarks.
FinAgent-RAG achieves 76.81-78.46% execution accuracy on financial QA benchmarks by combining contrastive retrieval, program-of-thought code generation, and adaptive strategy routing, outperforming baselines by 5.62-9.32 points.
XTR training does not improve retrieval effectiveness over ColBERT but enhances IVF engine efficiency by flattening token scores to produce more discriminative centroids.
A lightweight supervised router using frozen-LLM embeddings for memory admission decisions outperforms LLM-based memory managers in both F1 score and latency on the LoCoMo benchmark.
MiMIC mitigates visual modality collapse and semantic misalignment in universal multimodal retrieval via fusion-in-decoder architecture and robust single-modality training.
EvoRAG adds a feedback-driven backpropagation step that attributes response quality to individual knowledge-graph triplets and updates the graph to raise reasoning accuracy by 7.34 percent over prior KG-RAG methods.
Two-hop QA retrieval performance depends on whether the hop-2 entity is in the question or bridge passage, and a simple predicate-based router trained on one dataset transfers to improve R@5 on others.
ResearchEVO automates the discover-then-explain cycle by evolving algorithms via fitness-driven LLM co-evolution and generating grounded, anti-hallucination research papers through sentence-level RAG.
SelRoute routes queries to type-specific retrieval pipelines, achieving Recall@5 of 0.800 with a 109M model on LongMemEval_M and outperforming LLM-augmented baselines including a strong zero-ML lexical method.
ASTRA combines an eight-axis conceptual framework with text embeddings and unsupervised clustering to map and group 78 art-technology institutions into coherent thematic clusters.
LLM agents iteratively generate and optimize data processing strategies for fine-tuning, delivering over 80% win rates versus unprocessed data and 65% versus LLM-based AutoML baselines while cutting search time by up to 10x.