Picture for Yaojie Lu

Yaojie Lu

Search, Verify and Feedback: Towards Next Generation Post-training Paradigm of Foundation Models via Verifier Engineering

Add code
Nov 18, 2024
Viaarxiv icon

Transferable Post-training via Inverse Value Learning

Add code
Oct 28, 2024
Figure 1 for Transferable Post-training via Inverse Value Learning
Figure 2 for Transferable Post-training via Inverse Value Learning
Figure 3 for Transferable Post-training via Inverse Value Learning
Figure 4 for Transferable Post-training via Inverse Value Learning
Viaarxiv icon

Aligning Large Language Models via Self-Steering Optimization

Add code
Oct 22, 2024
Viaarxiv icon

A Unified View of Delta Parameter Editing in Post-Trained Large-Scale Models

Add code
Oct 17, 2024
Figure 1 for A Unified View of Delta Parameter Editing in Post-Trained Large-Scale Models
Figure 2 for A Unified View of Delta Parameter Editing in Post-Trained Large-Scale Models
Figure 3 for A Unified View of Delta Parameter Editing in Post-Trained Large-Scale Models
Figure 4 for A Unified View of Delta Parameter Editing in Post-Trained Large-Scale Models
Viaarxiv icon

StructRAG: Boosting Knowledge Intensive Reasoning of LLMs via Inference-time Hybrid Information Structurization

Add code
Oct 11, 2024
Figure 1 for StructRAG: Boosting Knowledge Intensive Reasoning of LLMs via Inference-time Hybrid Information Structurization
Figure 2 for StructRAG: Boosting Knowledge Intensive Reasoning of LLMs via Inference-time Hybrid Information Structurization
Figure 3 for StructRAG: Boosting Knowledge Intensive Reasoning of LLMs via Inference-time Hybrid Information Structurization
Figure 4 for StructRAG: Boosting Knowledge Intensive Reasoning of LLMs via Inference-time Hybrid Information Structurization
Viaarxiv icon

Multi-Facet Counterfactual Learning for Content Quality Evaluation

Add code
Oct 10, 2024
Viaarxiv icon

Seg2Act: Global Context-aware Action Generation for Document Logical Structuring

Add code
Oct 09, 2024
Figure 1 for Seg2Act: Global Context-aware Action Generation for Document Logical Structuring
Figure 2 for Seg2Act: Global Context-aware Action Generation for Document Logical Structuring
Figure 3 for Seg2Act: Global Context-aware Action Generation for Document Logical Structuring
Figure 4 for Seg2Act: Global Context-aware Action Generation for Document Logical Structuring
Viaarxiv icon

Rethinking Reward Model Evaluation: Are We Barking up the Wrong Tree?

Add code
Oct 08, 2024
Viaarxiv icon

READoc: A Unified Benchmark for Realistic Document Structured Extraction

Add code
Sep 08, 2024
Figure 1 for READoc: A Unified Benchmark for Realistic Document Structured Extraction
Figure 2 for READoc: A Unified Benchmark for Realistic Document Structured Extraction
Figure 3 for READoc: A Unified Benchmark for Realistic Document Structured Extraction
Figure 4 for READoc: A Unified Benchmark for Realistic Document Structured Extraction
Viaarxiv icon

Critic-CoT: Boosting the reasoning abilities of large language model via Chain-of-thoughts Critic

Add code
Aug 29, 2024
Viaarxiv icon