Lomonosov Moscow State University

Protein motif scaffolding usually fixes both where a motif sits along the chain and the relative pose of several rigid fragments, yet many applications supply neither. NomadFlow takes each motif group as a condition, holds each group's internal geometry exactly with a group projection, and lets the groups move independently through the first half of generation before locking them as conditioning to complete the scaffold. On 21 multi-group tasks adapted from Protpardelle, 19 reach unique success; against the same residue set fixed as one rigid body, the median ratio is 0.57 in yield and 1.00 in unique successes.

Protein motif scaffolding usually fixes both where a motif sits along the chain and the relative pose of several rigid fragments, yet many applications supply neither. NomadFlow takes each motif group as a condition, holds each group's internal geometry exactly with a group projection, and lets the groups move independently through the first half of generation before locking them as conditioning to complete the scaffold. On 21 multi-group tasks adapted from Protpardelle, 19 reach unique success; against the same residue set fixed as one rigid body, the median ratio is 0.57 in yield and 1.00 in unique successes.

Adaptive task allocation previously came with guarantees only when worker computation times stay fixed, yet in shared clusters resource contention, background jobs, or network congestion can turn a fast worker into a straggler and invalidate the allocation. Sliding Window Adaptive Task Allocation (SW-ATA) estimates each worker's delay mean from a recent window of observations and picks allocations from lower confidence bounds, so it can track fast workers while the delay distribution changes from round to round. On synthetic drifting-delay distributed stochastic optimization, it clearly beats the original adaptive allocation and uniform allocation, stays close to an oracle that knows the true means, and uses the least worker time.

Adaptive task allocation previously came with guarantees only when worker computation times stay fixed, yet in shared clusters resource contention, background jobs, or network congestion can turn a fast worker into a straggler and invalidate the allocation. Sliding Window Adaptive Task Allocation (SW-ATA) estimates each worker's delay mean from a recent window of observations and picks allocations from lower confidence bounds, so it can track fast workers while the delay distribution changes from round to round. On synthetic drifting-delay distributed stochastic optimization, it clearly beats the original adaptive allocation and uniform allocation, stays close to an oracle that knows the true means, and uses the least worker time.

CADFather is an autonomous, training-free agent in which a vision-language assistant decides which candidate CAD program to extend, which tool to invoke, and how many proposals to sample, unifying learned proposal generation, geometry-based algorithmic proposals, and parameter optimization in one CadQuery representation, and achieving zero invalidity with higher mean IoU and GMS and lower median Chamfer distance on the full DeepCAD, Fusion360, and MCB test sets.

CADFather is an autonomous, training-free agent in which a vision-language assistant decides which candidate CAD program to extend, which tool to invoke, and how many proposals to sample, unifying learned proposal generation, geometry-based algorithmic proposals, and parameter optimization in one CadQuery representation, and achieving zero invalidity with higher mean IoU and GMS and lower median Chamfer distance on the full DeepCAD, Fusion360, and MCB test sets.

For the Explainable Deepfake Detection Challenge on XPlainVerse, the authors present a modular detection-and-explanation system in which a detector fuses several DINOv3 backbones with Mesorch forensic features, a Grounding-DINO pipeline converts local artifact descriptions from explanations into weak patch-level pseudo-masks supervising an Artifact Evidence Map, and a local patch-level contrastive objective separates artifact from authenticity evidence, while class-conditional Qwen3-VL models generate complex explanations and a GRPO-optimized model simplifies them, reaching 0.9349 detection accuracy, a 0.5571 explanation score, and a 0.7456 final challenge score on the full test split.

For the Explainable Deepfake Detection Challenge on XPlainVerse, the authors present a modular detection-and-explanation system in which a detector fuses several DINOv3 backbones with Mesorch forensic features, a Grounding-DINO pipeline converts local artifact descriptions from explanations into weak patch-level pseudo-masks supervising an Artifact Evidence Map, and a local patch-level contrastive objective separates artifact from authenticity evidence, while class-conditional Qwen3-VL models generate complex explanations and a GRPO-optimized model simplifies them, reaching 0.9349 detection accuracy, a 0.5571 explanation score, and a 0.7456 final challenge score on the full test split.