{"total":11,"items":[{"citing_arxiv_id":"2605.22342","ref_index":8,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"4D-GSW: Kinematic-Aware Spatio-Temporal Consistent Watermarking for 4D Gaussian Splatting","primary_cat":"cs.CV","submitted_at":"2026-05-21T11:27:11+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"4D-GSW introduces a kinematic-aware spatio-temporal watermarking framework for 4D Gaussian Splatting that uses a Spatio-Temporal Curvature metric and HMM-MRF model to maintain consistency under attacks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.13835","ref_index":7,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Unlocking Patch-Level Features for CLIP-Based Class-Incremental Learning","primary_cat":"cs.CV","submitted_at":"2026-05-13T17:56:23+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"SPA unlocks patch-level features in CLIP for class-incremental learning via semantic-guided selection and optimal transport alignment with class descriptions, plus projectors and pseudo-feature replay to reduce forgetting.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.12241","ref_index":35,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Pretraining Strategies and Scaling for ECG Foundation Models: A Systematic Study","primary_cat":"eess.SP","submitted_at":"2026-05-12T15:10:15+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Contrastive predictive coding pretraining combined with structured state space models yields the strongest ECG foundation models, with continued gains from scaling data to 11 million samples.","context_count":1,"top_context_role":"method","top_context_polarity":"use_method","context_text":"CPC[ 22] learns representations by predicting future latent states from causal context via a contrastive objective, discriminating true future steps from within-sequence negatives over 5-second input segments.HuBERT++extends the speech HuBERT [ 23] approach by replacing offline hard k-means targets with EMA teacher-derived soft cluster assignments computed via Sinkhorn-Knopp optimal transport [35], encouraging balanced token participation and preventing representation collapse through online prototype updates, see App. S for details. We additionally evaluateECGFounder[ 8] andECG-JEPA[ 9], two of the best-performing external FMs in [20], as external FM reference points, and a supervised S4 model trained from scratch as a strong task- specific baseline."},{"citing_arxiv_id":"2605.11803","ref_index":7,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"OTT-Vid: Optimal Transport Temporal Token Compression for Video Large Language Models","primary_cat":"cs.CV","submitted_at":"2026-05-12T08:58:49+00:00","verdict":"UNVERDICTED","verdict_confidence":"MODERATE","novelty_score":6.0,"formal_verification":"none","one_line_summary":"OTT-Vid uses optimal transport with non-uniform token mass and locality-aware costs to dynamically allocate compression budgets across video frames, retaining 95.8% VQA and 73.9% VTG performance at 10% token retention.","context_count":1,"top_context_role":"method","top_context_polarity":"use_method","context_text":"This adaptive behavior is bounded by the 1/2 scaling, which confines αt to [1/2,1] so that semantic dissimilarity remains the dominant cost component while locality can contribute up to an equal weight in highly static frame pairs. Transport plan and compression difficulty.Given the mass vectors and the cost matrix, we solve an entropically regularized OT problem via Sinkhorn iterations [7]: Pt = argmin P∈Π(m t,mt+1) X i,j Pij Ct,ij −εH(P),(6) where Π(mt, mt+1) denotes all valid transport plans whose rows sum to mt and columns sum to mt+1, andH(P) =− P i,j Pij logP ij is the entropic regularizer that smooths the solution. The resulting transport plan Pt specifies an optimal coupling, where larger Pt,ij entries indicate high-priority candidates for compression."},{"citing_arxiv_id":"2605.05096","ref_index":2,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"CapsID: Soft-Routed Variable-Length Semantic IDs for Generative Recommendation","primary_cat":"cs.IR","submitted_at":"2026-05-06T16:33:13+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"CapsID uses probabilistic capsule routing and confidence-based termination to generate variable-length semantic IDs, improving recall by 9.6% over strong baselines with half the latency of dual-representation systems.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.01868","ref_index":31,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Robust Conditional Conformal Prediction via Branched Normalizing Flow","primary_cat":"cs.LG","submitted_at":"2026-05-03T13:29:26+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Branched Normalizing Flow improves conditional coverage robustness of conformal prediction under distribution shift by normalizing test inputs to the calibration distribution and mapping prediction sets back.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.27646","ref_index":60,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Benchmarking virtual cell models for in-the-wild perturbation response","primary_cat":"q-bio.CB","submitted_at":"2026-04-30T09:40:23+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"A new benchmarking framework shows virtual cell models overestimate performance on standard tests, drop sharply on unseen contexts and perturbations, and produce inconsistent rankings across metrics.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.23318","ref_index":6,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Hidden States Know Where Reasoning Diverges: Credit Assignment via Span-Level Wasserstein Distance","primary_cat":"cs.CL","submitted_at":"2026-04-25T14:11:23+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Span-level Wasserstein distances between hidden-state distributions of correct and incorrect rollouts provide a self-supervised signal to reweight advantages in GRPO, improving fine-grained credit assignment on math and code tasks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.09325","ref_index":11,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"A reduced-order model for parametrized Optimal Transport problems","primary_cat":"math.NA","submitted_at":"2026-04-10T13:50:56+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A reduced-order model for parametrized optimal transport problems using low-dimensional cone or subspace constraints and EIM-based error estimation.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2602.07633","ref_index":12,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Flow-Based Conformal Predictive Distributions","primary_cat":"stat.ML","submitted_at":"2026-02-07T17:26:50+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Differentiable nonconformity scores induce flows that sample conformal prediction set boundaries, and mixing flows across levels produces conformal predictive distributions whose quantiles match the sets.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2512.22597","ref_index":59,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Energy-Guided Generative Modeling for Low-Energy Molecular Structure Discovery","primary_cat":"cs.LG","submitted_at":"2025-12-27T14:00:22+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"EnFlow integrates flow-based conformer generation with energy landscape modeling to enable joint ensemble generation and ground-state identification using only 1-2 ODE steps.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null}],"limit":50,"offset":0}