2 papers
cs.AI2025
Effects of structure on reasoning in instance-level Self-Discover
Sachith Gunasekara, Yasiru Ratnayake
The drive for predictable LLM reasoning in their integration with compound systems has popularized structured outputs, yet concerns remain about performance trade-offs compared to…
cs.CL2024
OpenBezoar: Small, Cost-Effective and Open Models Trained on Mixes of Instruction Data
Chandeepa Dissanayake, Lahiru Lowe, Sachith Gunasekara +1
Instruction fine-tuning pretrained LLMs for diverse downstream tasks has demonstrated remarkable success and has captured the interest of both academics and practitioners. To ensur…