SWE-Fixer: open-source LLMs for effective and efficient Github Issues ResolutionOne of key LLMs applications for software engineering challenges, has emerged in resolving real-world tasks on GitHub by fixing code based…6d ago6d ago
AUTOPRESENT: LLM model for slide generation with results comparable to GPT-4oDesigning structured visuals such as presentation slides is essential for communicative needs, necessitating both content creation and…Jan 9Jan 9
CoT-based Synthesizer: inference scaling technique for enhancing LLM Performance through Answer…Current inference scaling methods, such as Self-consistency and Best-of-N, are effective in improving the accuracy of LLMs on complex…Jan 7Jan 7
TrustRAG: RAG Defense framework to counter attacks from malicious injected docsRAG systems are usually vulnerable to corpus poisoning attacks that can significantly degrade LLM performance through the injection of…Jan 41Jan 41
Confidence v.s.LLMs can correct their self-generated responses, but a decline in accuracy after self-correction is also witnessed. This paper[1] proposes…Dec 30, 2024Dec 30, 2024
SCOPE: KV Cache optimization framework for long-context generation in LLMsKey-Value (KV) cache has become a bottleneck of LLMs for long-context generation, with no current work focused on optimization for the…Dec 29, 2024Dec 29, 2024
Path-of-Thoughts: framework for robust relational reasoning with LLMsLLMs possess vast semantic knowledge but often struggle with complex reasoning tasks, particularly in relational reasoning problems such as…Dec 27, 2024Dec 27, 2024
Token-budget-aware LLM reasoning frameworkMethods like Chain-of-Thought (CoT) reasoning enhance LLM performance by decomposing problems into intermediate steps, however, they also…Dec 26, 2024Dec 26, 2024
RobustFT: framework for Robust Supervised Fine-tuning of LLMs under noisy responseSupervised fine-tuning (SFT) helps in adapting LLMs to specific domains or tasks, however do suffer from noise in data collection, thereby…Dec 24, 2024Dec 24, 2024
SPAR : Self-Play with Tree-Search Refinement to improve Instruction-Following in LLMsInstruction-following capabilities in LLMs requires model to recognize even the most subtle requirements in the instructions and…Dec 17, 2024Dec 17, 2024
