Topic
distillation
Commodities Ernakulam KVK Commissions 500 Kg Batch Lemongrass Distillation Unit to Boost Fringe Farming
The Krishi Vigyan Kendra (KVK) Ernakulam under ICAR-CMFRI has commissioned a 500 kg batch capacity lemongrass distillation unit at Vavely, Kottappady, in collaboration with CSIR-CIMAP under the Aroma Mission. The facility enables local processing of lemongrass into high-quality essential oil (80% citral content) and value-added products, aiming to provide sustainable livelihoods for farmers in forest-fringe areas affected by elephant raids.
New Framework Verifies Safety of Multi-Agent AI Communication for Autonomous Logistics
A new framework uses decision tree distillation to formally verify learned communication policies in multi-agent systems, targeting safety-critical autonomous logistics operations. The approach achieves 97.9% fidelity to neural policies and verifies 18 temporal logic properties with 88.9% satisfaction, including collision probabilities below 1% thresholds.
New Drift-RAE Method Distills Transformers Efficiently Using Representation Autoencoders
A new research paper proposes Drift-RAE, a method for distilling pretrained flow models in representation autoencoder latent spaces. It overcomes anisotropy and large curvature challenges, achieving 1.77 FID on ImageNet 256 with only 10,000 distillation steps, outperforming existing RAE distillation methods.
Open-SWE-Traces: 207K Multilingual Trajectories Set New Standard for Autonomous Software Engineering Agents
Researchers have released Open-SWE-Traces, a dataset of 207,489 software engineering agent trajectories spanning nine programming languages, sourced from 20,000 real-world pull requests. Fine-tuning on this data yields models that achieve state-of-the-art resolve rates on multiple SWE-bench benchmarks, advancing autonomous software engineering.
Mutual Distillation of Dual Foundation Models Achieves State-of-the-Art PET/CT Segmentation with Only 5 Labeled Cases
Researchers propose MuDuo, a mutual distillation framework that leverages two foundation models (SAM-Med3D for CT, SegAnyPET for PET) to distill knowledge into a lightweight student network for semi-supervised PET/CT segmentation. Achieving state-of-the-art performance on the AutoPET dataset with only 5 labeled cases, the approach eliminates manual prompts and maximizes unlabeled data utility.