ACL 2026 Highlights: Building More Adaptive and Trustworthy AI Systems
Correct answers are no longer enough. As LLMs become part of systems that use tools, memory, and evolving knowledge, we also need to understand how those components change model behavior.
DySECT: A Dynamic Self-Evolving Extraction and Curation Toolkit
Most LLM extraction systems treat each document in isolation, failing to retain and reuse knowledge acquired from previous extractions. That is a problem in domains where terminology changes, taxonomies evolve, and rare concepts matter. At ACL 2026 in San Diego, we introduce DySECT: a Dynamic Self-Evolving Extraction and Curation Toolkit. Two Ideas Behind DySECT Two […]
From Extraction to Adaptive Memory: Introducing DySECT
Information extraction is often treated as a one-pass prediction task: give a model a document, ask it to extract entities or relations, and the resulting knowledge is not reused to improve future extraction. That setup doesn’t work well when the domain keeps changing. In medical, legal, HR, scientific, and enterprise knowledge settings, terminology evolves, rare […]
When Strong LLMs Fail at Tool Use: What FuncBenchGen Reveals
FuncBenchGen provides a contamination-free framework for systematically stress-testing multi-step tool reasoning and exposing hidden failure modes.
Optimizing Human-AI Collaboration: A Conversation with Stanford’s Diyi Yang
Stanford’s Diyi Yang on optimizing human-AI collaboration: how General User Models and Next Action Prediction help AI adapt to people, not just respond.
Tool-Induced Myopia in LLMs: What Tool Use Means for AI Reasoning
Megagon Labs’ ACL 2026 paper explores Tool-Induced Myopia, showing how tool use can improve LLM accuracy while degrading AI reasoning quality.
Memory-Augmented Learning for LLM Agents Without Fine-Tuning
LLM agents can now learn continuously from experience, without a single parameter update. Our latest research at Megagon Labs introduces a memory-driven framework that enables agentic systems to improve from feedback without the cost, inflexibility, and opacity of fine-tuning. Why Fine-Tuning Falls Short Traditional approaches to improving LLM performance rely on parameter updates, which are […]
Trends in Agentic AI and LLM Systems at EACL 2026
Trends in Agentic AI & LLM Systems Trends in Agentic AI & LLM Systems The 19th conference of the European Chapter of the Association for Computational Linguistics (EACL) took place in Rabat, Morocco, from March 24 to 29, 2026. It brought together researchers, practitioners, and industry leaders from around the world. This year marked a […]
Representation Matters More Than You Think in LLM Systems
How Data Representation Shapes Compound AI Systems
7 AI Research Internship Tips for Future Jobs in AI
Advice from Megagon Labs for students ready to go beyond a resume line item.
