home.social

#aiplanning — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #aiplanning, aggregated by home.social.

  1. OmniEVA: Bridging the 2D–3D Gap in Embodied AI

    New paper introduces OmniEVA, a versatile embodied planner that pushes the boundaries of multimodal large language models (MLLMs) for robotics and spatial reasoning.

    Results: OmniEVA achieves state-of-the-art performance across 2D/3D reasoning benchmarks and outperforms existing models in object navigation tasks.

    Paper: arxiv.org/pdf/2509.09332v1
    Project: omnieva.github.io/

    #EmbodiedAI #Robotics #LLM #MLLM #3DVision #AIResearch #AIPlanning

  2. OmniEVA: Bridging the 2D–3D Gap in Embodied AI

    New paper introduces OmniEVA, a versatile embodied planner that pushes the boundaries of multimodal large language models (MLLMs) for robotics and spatial reasoning.

    Results: OmniEVA achieves state-of-the-art performance across 2D/3D reasoning benchmarks and outperforms existing models in object navigation tasks.

    Paper: arxiv.org/pdf/2509.09332v1
    Project: omnieva.github.io/

    #EmbodiedAI #Robotics #LLM #MLLM #3DVision #AIResearch #AIPlanning

  3. If you need efficient constraint solving with long-term support, Timefold is the future. Less waiting, more optimizing. 🚀 #Optimization #Timefold #AIPlanning

    Having used OptaPlanner in multiple projects and now testing Timefold, I can confidently say it outperforms OptaPlanner in every way. The devs are active and receptive while Red Hat seems to have quietly shelved OptaPlanner moving it to the KIE / Apache team who do not offer support.

    timefold.ai/

  4. If you need efficient constraint solving with long-term support, Timefold is the future. Less waiting, more optimizing. 🚀 #Optimization #Timefold #AIPlanning

    Having used OptaPlanner in multiple projects and now testing Timefold, I can confidently say it outperforms OptaPlanner in every way. The devs are active and receptive while Red Hat seems to have quietly shelved OptaPlanner moving it to the KIE / Apache team who do not offer support.

    timefold.ai/

  5. #AI #ML research/publishing operates in silos - to the detriment of making progress.

    Our #IJCAI submission on #OpenWorldLearning #OWL was rejected for good and bad reasons.

    The bad reason: "this is not just planning but also something similar to reinforcement learning".

    Guess what - that is the point of our research! We are trying to close the gap between designed #AIPlanning systems and adaptive #Learning systems. It is a super-hard gap to push #AI #ML algorithmic research in.

  6. #AI #ML research/publishing operates in silos - to the detriment of making progress.

    Our #IJCAI submission on #OpenWorldLearning #OWL was rejected for good and bad reasons.

    The bad reason: "this is not just planning but also something similar to reinforcement learning".

    Guess what - that is the point of our research! We are trying to close the gap between designed #AIPlanning systems and adaptive #Learning systems. It is a super-hard gap to push #AI #ML algorithmic research in.

  7. Common wisdom in #AI and #ML is that #AIPlanning methods cannot deal with continuous state and action spaces.

    Subverting these expectations - presenting our recent paper at #ICAPS23 on how a planning agent can play #AngryBirds!

    And, no #DL #DQN systems cannot play these games yet AND take so much data to learn to play a single level.

    #Planning #Reasoning #KRR FTW!!

    arxiv.org/abs/2303.16967

  8. Common wisdom in #AI and #ML is that #AIPlanning methods cannot deal with continuous state and action spaces.

    Subverting these expectations - presenting our recent paper at #ICAPS23 on how a planning agent can play #AngryBirds!

    And, no #DL #DQN systems cannot play these games yet AND take so much data to learn to play a single level.

    #Planning #Reasoning #KRR FTW!!

    arxiv.org/abs/2303.16967

  9. At the meeting of the French #robotics research group #GdRRobotique, I liked Nick Hawes’ STRAND project about long-term #HRI. For 3 months they had an interactive robot:
    - explore to learn times and locations more eager to lead to interactions into an #MDP
    - assess the risk of requiring an interaction to move on, e.g. when it needed someone to open a door
    - exploit by maximising the number of interactions per day, using a tool called PRISM #AIPlanning
    arxiv.org/pdf/1604.04384.pdf