Prevent Safety Drift in LLMs with CWAC Method
Discover how Coupled Weight and Activation Constraints (CWAC) prevent safety drift in large language models while maintaining accuracy during fine-tuning.
ReflectCAP: Advanced Image Captioning with Reflective Memory
Discover ReflectCAP, a cutting-edge image captioning method using reflective memory for improved factuality and detailed coverage across top LVLMs.
MultiDocFusion: Advanced Chunking for Long Industrial Docs
Discover MultiDocFusion, a multimodal chunking pipeline that boosts RAG accuracy on long industrial documents through hierarchical parsing and OCR.
Frontier-Eng: Benchmarking AI in Real-World Engineering Tasks
Discover Frontier-Eng, a benchmark for evaluating self-evolving AI agents on real-world engineering tasks using generative optimization techniques.
GAM: Hierarchical Graph Memory Boosts LLM Agent Accuracy
Discover GAM, a hierarchical graph-based memory framework enhancing long-term reasoning and context retention in Large Language Model agents.