Monte Library 私人知识库 Library EN

LLM Wiki

方法与证据比较

Scope
cnben
Release
cnbben-v2-324babf8774f
Source digest
324babf8774f7086d55800c61c9ad226096521b120767bfdb25d14de00ee9d3f
Schema
1
Published page root
llm_wiki/published
Manifest
llm_wiki/reports/releases/cnbben-v2-324babf8774f/release.json
Page kind
comparison
Page source path
llm_wiki/published/comparisons/methods.md

Content

page_id: method-comparison kind: comparison language: zh-CN claim_id: clm-method-comparison-001 claim_status: synthesis

方法与证据比较

研究对象 问题 方法 数据集 指标 局限 证据
BELB Read Deep 待补证据 待补证据 待补证据 待补证据 待补证据 卡片
BeLink Read Deep Biomedical Entity Linking (BEL) with large language models (LLMs) remains computationally inefficient and challenging to deploy in practical settings.<br><sub>证据：papers/CNBEN/BeLink/BeLink Read Deep.md（line:38）</sub> A BEL pipeline using re-ranking stage through generative models<br><sub>证据：papers/CNBEN/BeLink/BeLink Read Deep.md（line:37）</sub> 待补证据 待补证据 待补证据 卡片
经典rerank Read Deep 待补证据 待补证据 待补证据 待补证据 待补证据 卡片
BGE-M3 Read Deep 待补证据 M3-Embedding, which is distinguished for its versatility in Multi-Linguality,<br><sub>证据：papers/CNBEN/BGE-M3/BGE-M3 Read Deep.md（line:41）</sub> 待补证据 待补证据 待补证据 卡片
BioELX Read Deep However, expert-annotated training data for BEL are costly, especially for low-resource languages. Moreover, many cross-lingual BEL systems rely on SapBERT-based retrievers trained on predominantly English aliases in the KB, leading to poor generalization to unseen non-English mentions and limited context-aware disambiguation.<br><sub>证据：papers/CNBEN/BioELX/BioELX Read Deep.md（line:39）</sub> BioELX, a two-stage cross-lingual BEL framework that requires no task-specific annotated training corpora.<br><sub>证据：papers/CNBEN/BioELX/BioELX Read Deep.md（line:38）</sub> 待补证据 待补证据 待补证据 卡片
Biomedical Entity Representations with Synonym Marginalization Solve problems of the incompleteness of provided synonyms and numerous variations in their surface forms.<br><sub>证据：papers/CNBEN/BioSyn/BioSyn.md（line:21）</sub> Solve problems by using a model-based candidate selection and maximize the marginal likelihood of the synonyms present in top candidates<br><sub>证据：papers/CNBEN/BioSyn/BioSyn.md（line:22）</sub> NCBI Disease Corpus; Biocreative V CDR; TAC2017ADR 待补证据 待补证据 卡片
CBLUE Abstract Summary 待补证据 待补证据 待补证据 待补证据 待补证据 卡片
CoRTEx 待补证据 Turns to encoder terms into low-dimensional embeddings and performs term clustering or entity linking based on the embedding similarity.<br><sub>证据：papers/CNBEN/CoRTEx/CoRTEx.md（line:18）</sub> 待补证据 待补证据 待补证据 卡片
Qwen3 Read Deep Text embedding and reranking capabilities<br><sub>证据：papers/CNBEN/Qwen3/Qwen3 Read Deep.md（line:37）</sub> Qwen3 Embedding series are a series of advanced models in text embedding and reranking capabilities, built upon the Qwen3 foundation models.<br><sub>证据：papers/CNBEN/Qwen3/Qwen3 Read Deep.md（line:37）</sub> 待补证据 待补证据 待补证据 卡片
医学同义词模型 Read Wide 待补证据 待补证据 待补证据 待补证据 待补证据 卡片
Self-Alignment Pretraining for Biomedical Entity Representations SapBERT is a pretraining model that self-aligns the representation space of biomedical entites<br><sub>证据：papers/CNBEN/SapBERT 1/SapBERT.md（line:22）</sub> Solves problem that accurately capturing fine-grained semantic relationships in the biomedical domain remains a challenge.<br><sub>证据：papers/CNBEN/SapBERT 1/SapBERT.md（line:23）</sub> 待补证据 待补证据 待补证据 卡片

所有“待补证据”均表示原始笔记当前未覆盖该字段。

Claims

Claim metadata

Switch to English