The open question is how to get AI systems (LLMs or otherwise) to induce novel causal relationships from experience.
AGENTS
-
GAIA: Benchmark for General AI Assistants Capabilities
By
–
8/ GAIA – a benchmark for general AI assistants consisting of real-world questions that require a set of fundamental abilities such as reasoning, multimodal handling, web browsing, and generally tool-use proficiency.
-
LLMs as Collaborative Agents for Medical Reasoning
By
–
9/ LLMs as Collaborators for Medical Reasoning – proposes a collaborative multi-round framework for the medical domain that leverages role-playing LLM-based agents to enhance LLM proficiency and reasoning capabilities.
-
Chain-of-Thought Reasoning and Language Agents Guide
By
–
7/ The Hitchhiker’s Guide From Chain-of-Thought Reasoning to Language Agents – summary of CoT reasoning, foundational mechanics underpinning CoT techniques, and their application to language agent frameworks.
-
Top ML Papers: Mirasol3B, System 2 Attention, Speculative Sampling
By
–
Top ML Papers of the Week (Nov 20 – Nov 26): – Mirasol3B
– System 2 Attention
– Parallel Speculative Sampling
– Advancing Long-Context LLMs
– Teaching Small LMs To Reason
– LLMs as Collaborators for Medical Reasoning
… -
Instrumental Objectives and Low-Level Guardrails in AI Design
By
–
Right, the old "instrumental objective" story.
You can have low-level guardrails against bad effects of instrumental sub-goals.
The question here is not "can you come up with a way that this could go wrong?", but rather "is there a way to do it right?" It's like turbojet design. -
Tool Use Methodology: Retrieval Approach in AI Systems
By
–
There’s a paper describing this approach, for one particular tool (retrieval); idea is the same however for other tools https://
arxiv.org/abs/2310.11511 -
Alignment Concerns with Internal Health Reward System Design
By
–
The first comment (kudos for open review) links to a post that says some of this at greater length, but to repeat my own reaction: "There's nothing in there about alignment. The proposed motivational system is internal-system-health reward with nothing about caring for humans."