HindSearch introduces trajectory-level hindsight critique with on-policy distillation, boosting average EM to 39.4% on seven QA benchmarks, outperforming prior methods.

Haowei Liu, Jiamian Wang, Hsin-Tai Wu et al.

Proposed D²-4DGS fuses dual-depth priors for sparse-camera 4D Gaussian scene synthesis, improving PSNR by 1.33dB.

Jijian Zhao

STAR-VLM uses nuScenes radar supervision to train VLMs, reaching 0.94 motion accuracy and 1.20 m/s radial-velocity MAE.

Pou-Chun Kung, Aryaman Rao, Utkrisht Sahai et al.

Pre-trained detection transformers (e.g., DETR) encode depth and 3D position info in object embeddings without explicit 3D supervision, as shown by probing experiments.

Robin Kim, Colin Samplawski, Benjamin M. Marlin

DynamicManip synthesizes diverse dynamic demonstrations from a single static example, boosting data efficiency and real-time response in robot manipulation.

Haoran Liao, Pengyue Wang, Shuoyu Chen et al.

PNPO replaces cumulative importance ratios with prefix geometric mean, improving off-policy correction in large language model RL.

Wenhao Zhang, Yibo Xie, Rui Wang et al.

KoVRE model enhances Korean document retrieval efficiency using bilingual supervision and hard-negative mining.

Yongbin Choi, Gyuho Shim, Youngjoon Jang

Hybrid AI architecture combining LLMs and DCR graphs enhances explainability and accuracy of eGovernment conversational agents.

Ilias Chalkidis, Vlad Paul Cosma, Søren Debois et al.

AlphaG-OPD enhances reliability in symbolic factor discovery via structured on-policy distillation, showing strong cross-market performance.

Yaoyu Su

Using data augmentation as a framework to model hippocampal contributions to generalization, proposing online and offline strategies.

Tyler Bonnen, Andrew Kyle Lampinen

VaRS-Doc enhances visual document retrieval by diversifying document representations via latent self-probing, achieving state-of-the-art performance.

Haocheng Wang, Tongkun Guan, Wei Shen et al.

NovaCov uses set-wise selection with a recency-weighted reference bank, retaining 99.6% accuracy and reducing latency by 46% in streaming video token compression.

Moxu Duan, Jingwen Fu, Yuwang Wang

OC-VLA++ enhances viewpoint robustness in monocular robotic manipulation via geometry-guided cross-view supervision and action equivariance, achieving 8-15% success rate improvements under large camera shifts.

Tianyi Zhang, Ziyang Gong, Zhenjie Yang et al.

Proposes Cloud-ScPO, leveraging hidden-state geometry for semi-supervised preference optimization, achieving up to 4.49% accuracy gain on GSM8K.

Yuzhou Liu, Xiyang Hu

Proposes human-AI system (HAS) framework for scientific teams, emphasizing dynamic feedback and collaboration to mitigate risks and enhance innovation.

Patrick Emami, Sameera Horawalavithana, Truc Nguyen et al.

PMMC shifts multimodal long-term memory from query-time retrieval to consolidation, improving answer accuracy and efficiency by 3.2-4.0 points on benchmarks.

Jingyu Sun, Yan Lin, Yuyang Xue et al.

Introduces Tevatron 3.0 with Megatron backend, enabling efficient training of billion-parameter MoE rerankers, surpassing FSDP in speed and scale.

Zhichao Xu, Xueguang Ma, Shengyao Zhuang et al.

AOSpec jointly speculates actions and observations, reducing end-to-end latency by 11.8%-32.5% using lossless verification and EV-based observation drafting.

Hao Mark Chen, Jinnan Guo, Wayne Luk et al.

Pruned BPE introduces post-training visibility pruning and token reallocation, reducing encoded length by ~0.3% while preserving vocabulary size.

Kenny Shao

Introducing 'isotropy cliffs' as a geometric signature of decision-making in LLMs; strong correlation (r=0.84) with downstream accuracy.

Okan S. Coskun, Florian Rottach, Carsten Eickhoff et al.