Teaching robots to ask for clarification in 3D

Kari Jaaskelainen

12 Jan 2026 — 1 min read

When robots should ask: Which one?

In safety-critical places like operating rooms, a vague command like "Pass me the vial" can be dangerous. This paper introduces a simple idea with big impact: teach AI to detect when an instruction is ambiguous in a 3D scene and pause to ask for clarification.

New task: Open-Vocabulary 3D Instruction Ambiguity Detection — decide if a command has exactly one clear target in a scene.
New dataset: Ambi3D with 700+ diverse scenes and ~22k instructions to stress-test models.
Key finding: Today’s leading 3D LLMs often miss ambiguity.
New method: AmbiVer, a two-stage system that gathers visual evidence from multiple views and uses it to judge clarity more reliably.

Why it matters: More cautious, trustworthy assistants — from hospitals and labs to warehouses and homes.

Read more: https://arxiv.org/abs/2601.05991 and project/code: https://jiayuding031020.github.io/ambi3d/

Paper: https://arxiv.org/abs/2601.05991v1

Register: https://www.AiFeta.com

#AI #Robotics #Safety #ComputerVision #3D #LLM #VLM #HRI #EmbodiedAI

Meet GRACE: a moral governor for safer, more transparent AI

AI agents are getting powerful—so how do we make sure they do the right thing, not just the effective thing? Meet GRACE, a reason-based moral governor that keeps AI behavior aligned with human norms by separating moral reasoning from goal-driven decision-making. * Moral Module: uses deontic logic and explicit reasons

Why some fine-tuned LLMs miss phishing—and how to fix it

Not all fine-tuned LLMs spot phishing equally. A new study tests Llama 3.1 8B, Gemma 2 9B, and Mistral on high-stakes phishing detection—and uses SHAP and mechanistic interpretability to reveal why models do (or don’t) generalize. * Architecture × data diversity matters: Gemma 2 9B hits state-of-the-art performance (F1

AI that explains itself—by following the science

AI that explains itself—by following the science Black-box AI can be powerful, but it often can’t tell us why it made a decision. Concept Bottleneck Models (CBMs) try to fix that by predicting human-understandable concepts first, then the final answer. The catch: standard CBMs ignore domain-specific cause-and-effect and

Meet GenomAgent: A Team of AI Specialists for Smarter Genomics Q&A

TL;DR: Finding reliable answers in genomics is hard. GenomAgent turns one big AI into a coordinating team of specialists—and beats the current leader by 12% on a key benchmark. Genomic facts live across many databases. Standard chatbots struggle because they can’t flexibly query those sources. GeneGPT added