Masked generation, confidence-based unmasking and parallel decoding. Compare which tokens are generated first and at what computational cost.
Pengcheng Huang
chengpingan
·
AI & ML interests
AI & Agent
Recent Activity
new activity about 15 hours ago
evaleval/EEE_datastore:[Submission] Paper-reported comparisons for translation, retrieval, diffusion decoding and LLM agents updated a dataset 1 day ago
chengpingan/lexically-constrained-translation-results published a dataset 1 day ago
chengpingan/lexically-constrained-translation-resultsOrganizations
None yet
Terminology-constrained machine translation
Compare decoding with lexical constraints to revising missing glossary terms after translation. Track term coverage and translation quality.
-
Lexically Constrained Decoding for Sequence Generation Using Grid Beam Search
Paper • 1704.07138 • Published • 1 -
Fast Lexically Constrained Decoding with Dynamic Beam Allocation for Neural Machine Translation
Paper • 1804.06609 • Published • 1 -
Translate-and-Revise: Boosting Large Language Models for Constrained Translation
Paper • 2407.13164 • Published
Decoding strategies for diffusion language models
Masked generation, confidence-based unmasking and parallel decoding. Compare which tokens are generated first and at what computational cost.
Terminology-constrained machine translation
Compare decoding with lexical constraints to revising missing glossary terms after translation. Track term coverage and translation quality.
-
Lexically Constrained Decoding for Sequence Generation Using Grid Beam Search
Paper • 1704.07138 • Published • 1 -
Fast Lexically Constrained Decoding with Dynamic Beam Allocation for Neural Machine Translation
Paper • 1804.06609 • Published • 1 -
Translate-and-Revise: Boosting Large Language Models for Constrained Translation
Paper • 2407.13164 • Published
spaces 7
Running
Unmasking Order in Diffusion Language Models
🧩
How diffusion language models choose what to unmask next
Running
Query Expansion for Dense Retrieval
📖
Compare LLM query expansion and joint retriever training
Running
Terminology-Constrained Translation
📖
Check glossary terms and compare draft and revision
Running
Edge-Cloud Agent Routing
📖
Compare routing, delegation, cost and privacy exposure
Running
LLM Tool Use Generalization
📖
Test how tool trials help agents use unfamiliar tools
Running
Long-Context Preference Optimization
📖
Explore bandit chunk sampling for long-context DPO
datasets 10
chengpingan/lexically-constrained-translation-results
Viewer • Updated • 16 • 14
chengpingan/masked-diffusion-decoder-results
Viewer • Updated • 19 • 13
chengpingan/CoFaithfulQA
Viewer • Updated • 8.53k • 326
chengpingan/ExpandR_llm_training_data
Viewer • Updated • 30k • 44 • 1
chengpingan/LLM-QE-DPO-Training-Data
Viewer • Updated • 30k • 32
chengpingan/ExpandR_retriever_training_data
Viewer • Updated • 709k • 52 • 1
chengpingan/LLM-QE-Retriever-Training-Data
Viewer • Updated • 709k • 29
chengpingan/ParamMute-Training-Data
Viewer • Updated • 32.6k • 45 • 2
chengpingan/EC2-Bench
Viewer • Updated • 128 • 2.75k • 1
chengpingan/wildclaw_trej
Updated • 8