Large Language Model
Can LLMs Solve Molecule Puzzles A Benchmark for Molecular Structure Elucidation
LLMs with improved chemical reasoning capabilities in the other two sub-tasks.
Semantic Density: Uncertainty Quantification for Large Language Models through Confidence Measurement in Semantic Space
A new framework is proposed in this paper to address these issues.
InfoRM: Mitigating Reward Hacking in RLHF via Information-Theoretic Reward Modeling Y uchun Miao
With the advent of large language models (LLMs), reinforcement learning from human feedback (RLHF) has emerged as a pivotal technological paradigm to align models' behaviors with human values [
Meta-Controller: Few-Shot Imitation of Unseen Embodiments and Tasks in Continuous Control
Robot embodiments vary widely in their morphological ( e.g., number and connectivity
Copilot bug allows 'AI' to read confidential Outlook emails
PCWorld reports on a critical Microsoft Copilot bug (CW1226324) that allows the AI to scan and summarize confidential Outlook emails, bypassing privacy protections. This vulnerability affects Microsoft 365 accounts and compromises sensitive data like contracts and medical information stored in Sent and Drafts folders. Microsoft is rolling out a fix, but the timeline remains unclear, raising significant concerns about AI reliability and data privacy protection. For all its supposed intelligence, "AI" seems to make a lot of stupid mistakes--for example, scanning and summarizing emails marked "confidential" in Microsoft Outlook.
A Practitioner's Guide to Continual Multimodal Pretraining
However, practical model deployment often operates in the gap between these two limit cases, as real-world applications demand adaptation to specific subdomains, tasks or concepts -- spread over the entire, varying life cycle of a model.
Google DeepMind wants to know if chatbots are just virtue signaling
Google DeepMind is calling for the moral behavior of large language models--such as what they do when called on to act as companions, therapists, medical advisors, and so on--to be scrutinized with the same kind of rigor as their ability to code or do math . As LLMs improve, people are asking them to play more and more sensitive roles in their lives. Agents are starting to take actions on people's behalf. LLMs may be able to influence human decision-making . And yet nobody knows how trustworthy this technology really is at such tasks. With coding and math, you have clear-cut, correct answers that you can check, William Isaac, a research scientist at Google DeepMind, told me when I met him and Julia Haas, a fellow research scientist at the firm, for an exclusive preview of their work, which is published in today. That's not the case for moral questions, which typically have a range of acceptable answers: "Morality is an important capability but hard to evaluate," says Isaac. "In the moral domain, there's no right and wrong," adds Haas.