Pinned Tweet
Excited to share our work to appear at #CVPR2025! This work in collaboration with @jacob__krantz and @stefmlee investigates if visual representations (imaginations) of natural language instructions can improve performance of vision-and-language navigation (VLN) agents.
TL;DR-yes
Akhil retweeted
We are grateful to all of the 17,491 reviewers who helped make #CVPR2026 possible. We are especially pleased to recognize the following Outstanding Reviewers, whose high-quality reviews (as judged by their Area Chairs) placed them among the top 5% of reviewers.
Akhil retweeted
Introducing PEGrad, a hyperparameter-free method for energy-efficient robot control. By projecting energy gradients orthogonal to task rewards, it cuts energy use by up to 64% without hurting performance. Demonstrated on DM-Control, HumanoidBench, and real Unitree GO2 robots,
Key takeaways:
- Extends robot battery life and safety.
- Removes the need for tedious hyperparameter tuning between energy and task rewards.
- Brings us closer to sustainable, deployable energy-efficient RL robots.
Website: pvskand.github.io/projects/P…
Paper: Non-conflicting Energy Minimization in Reinforcement Learning based Robot Control - arxiv.org/abs/2509.01765
Happy to share that our collaborative work, “Harnessing Input-Adaptive Inference for Efficient VLN” has been accepted to #ICCV2025 ! 🤖
Paper: arxiv.org/pdf/2508.09262
(contd.)
Combined, these methods achieve about 56% average reduction in computation, with only about 12% average drop in success rate (SR) across standard VLN benchmarks—and even higher savings (~87–90%) in continuous VLN environments with modest performance trade-offs.
Happy to be recognized as an outstanding reviewer at #CVPR2025. :)
Behind every great conference is a team of dedicated reviewers. Congratulations to this year’s #CVPR2025 Outstanding Reviewers!
cvpr.thecvf.com/Conferences/…
We use a language-conditioned pre-trained diffusion model to generate imaginations by leveraging its strong priors. The imagination representations are aligned to corresponding noun-phrases from instructions using an auxiliary alignment loss.
Finally, they are encoded along with other modalities in a cross-modal encoder to sample agent actions. We observe performance improvements across agents and natural language instruction granularities. For more details, our work is available at akhilperincherry.com/VLN-Ima….