Hamid Rezatofighi

RT @zhixi_cai: 🎉🚁 I’m thrilled to share that our paper “NEUSIS: A Compositional Neuro-Symbolic Framework for Autonomous UAV Search Missions…

RT @zhixi_cai: 🎉🚀 I’m thrilled to share that our paper “NAVER: A Neuro-Symbolic Compositional Automaton for Visual Grounding with Explicit…

Two papers from #VL4AI @MonashUni accepted to #ICCV2025 1.NAVER: Combines VLM, logic & automata. SOTA on 6 visual grounding benchmarks. 2.DWIM: Train LLMs for visual reasoning via agentic neuro-symbolic learning. Strong generalisation, SOTA results #NeuroSymbolic #VisualReasoning https://t.co/WLNIfMLCNl

Very well deserved works. Both on 3D Vision

Great to see Neuro-symbolic AI gaining momentum in computer vision, especially in the context of robotics and embodied AI problems. Inspiring talk by @DrChenWang at the @CVPR AC Workshop! #CVPR2025 #EmbodiedAI #NeuroSymbolic #ComputerVision #AIResearch https://t.co/nBwISHRuCt

Stats on the accepted papers in @CVPR by the primary topics, presented by #CVPR2025 PCs in AC workshop. Generative models is still most trending topics by large margin. Sorry to see CV in robotics to be this low. https://t.co/6YgjkNYP5S

Heading to Nashville for #CVPR2025, Vanderbilt visit & DARPA PI meeting! Presenting: 🔹 NAVER (Neuro-symbolic reasoning) – June 9 🔹 DrVideo (Long video understanding) – June 14 🔹 HARNESS talks @ Vanderbilt + DARPA – June 15–26 DM if you’re around! #AI #Robotics #NeuroSymbolicAI https://t.co/lXPkGU7jmw

5-DWIM: Towards Tool-aware Visual Reasoning via Discrepancy-aware Workflow Generation & Instruct-Masking Tuning. Paper: https://t.co/sj1xxBFvzV

4- NAVER: A Neuro-Symbolic Compositional Automaton for Visual Grounding with Explicit Logic Reasoning, Arxiv 2025. Paper: https://t.co/5uuXAu3B6e

3- HYDRA: A Hyper Agent for Dynamic Compositional Visual Reasoning, ECCV 2024. Paper: https://t.co/ugrcnPBly5

2- 2- Hier-SLAM & Hier-SLAM++: Neuro-Symbolic Semantic SLAM with a Hierarchically Categorical Gaussian Splatting, Arxiv 2025 (earlier version in ICRA 2025). Paper: https://t.co/GdBWw7Q8ah

Some of our recent papers which will be covered in HARNESS talk: 1- NEUSIS: A Compositional Neuro-Symbolic Framework for Autonomous Perception, Reasoning, and Planning in Complex UAV Search Missions, Arxiv 2025. Paper: https://t.co/WHz6twRDYa

A great work by ByteDance with an open sourced code. Congratulations to Chenhui Guo from @MonashInfotech for his involvement in this work.

The source code for our Hier-SLAM has been released. #SLAM #Robotics @ieee_ras_icra

RT @physical_int: We got a robot to clean up homes that were never seen in its training data! Our new model, π-0.5, aims to tackle open-wor…

Dear @ICCVConference PCs, how as an ICCV25 area chair I should complete reviewer registration? I receive quite few warning emails for this. Could you please exclude ICCV25 ACs from your automatic warning emails?

RT @MonashInfotech: @MonashUni has soared 24 places to rank 63rd among universities worldwide in the prestigious @timeshighered World Uni R…

Two papers from my team at #VL4AI @MonashInfotech, accepted to #ICRA2025 1. Hier-SLAM: Scaling-up Semantics in SLAM with a Hierarchically Categorical Gaussian Splatting 2. Social-MAE: Social Masked Autoencoder for Multi-person Motion Representation Learning #AI #Robotics https://t.co/eaXcbprKUx

We are hiring a Postdoc in Embodied AI or 3D Vision, or Robot Vision. The candidate must be already in Australia or hold a valid visa & be available to start immediately. Strong research expertise & publication records are required. Interested, Please email me your CV. #3DVision

RT @CVPR: If you happen to miss the #CVPR2025 review deadline (though it’s not ideal), inform your area chair IMMEDIATELY and provide a cle…