LLM system capabilities depend not only on model parameters but also on the executable harnesses that organize information acquisition, tool use, reasoning, and memory. Recursive harness evolution can automate their design, but how it can reliably discover generalizable systems from limited, noisy, and repeatedly reuse...
Zi-Hao Lin, Hao-Yu Dong, Hao-Bo Xu et al.· 1 citation
Diffusion-based driving planners capture diverse behaviors but can generate unsafe trajectories under distribution shift. We propose BridgeGuard, a safety-constrained diffusion planning method that progressively strengthens a constraint term during denoising to drive intermediate trajectories toward a scene-dependent s...
Zhen-Jun Qiu, Jia-Ning Huang, Dongfang Liu et al.· 0 citations
This work presents a diagnostic and optimization framework grounded in a key empirical finding: the value of tokens within a CoT reasoning sequence is highly non-uniform, and this non-uniformity can be effectively characterized by token-level log probability signals.
Run-Jia Zeng, Hang Hua, Yiyang Liu et al.· arXiv.org· 1 citation
A novel Reliability-Aware framework for MMEA (RA-MMEA), which assesses visual reliability and adaptively improves unreliable visual representations for robust entity alignment and enables more reliable visual representations for modality fusion, thereby improving performance.
Chen-Xiao Li, Yun-He Feng, Dongfang Liu et al.· 0 citations
VLA-Scope is introduced, a two-stage framework that combines input-shift characterization with execution history to predict failure during OOD rollouts and achieves a higher ROC-AUC than the evaluated ActProbe and SAFE-MLP baselines.