Agent4Spec is presented, an autonomous multi-agent framework featuring two distinct pipelines: a validation pipeline that detects and self-corrects OpenAPI syntax errors, and a market analysis pipeline that conducts competitive research, identifies functional gaps against industry leaders, and automatically generates m...
Yong-Jian Tang, C. Moraru, Doruk Tuncel et al.· 0 citations
In-context-learning Tabular Foundation Models (TFMs) are uniquely suited to scenarios with limited resources or privacy constraints, as they solve new tasks without additional training. However, their architecture consists of multiple transformer blocks, resulting in high computational and memory footprints. Meanwhile,...
Recent advances have significantly improved the performance of Abstract Meaning Representation (AMR) parsers, yet determining when their outputs can be trusted remains largely unexplored. This work presents the first study of uncertainty estimation for AMR parsers in a black-box setting using top-k beam-search candidat...
Anuj Kumar, Shintaro Ozaki, Yamuna Prasad et al.· 0 citations
Generative modeling for discrete data has been significantly advanced by learning continuous vector fields on the statistical manifolds to transport simple priors to target discrete distributions. However, because these continuous-time flows are not constrained to follow geodesics on the statistical manifold, thereby r...
Xiao-Huan Jia, Renzhe Xu, Xiao Wang et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Actions specify what a robot should execute, but do not describe what those commands will achieve. Vision-language-action (VLA) policies nevertheless learn primarily by predicting action chunks from images, language, and state. We introduce Action--Future Coupled Effect Learning (AFCE), which makes a future-grounded ph...
Min Lin, Bo-Kai Zhou, Jian-Qi Lin et al.· 0 citations
Surprisingly, frozen recurrent networks support recall over hundreds of steps and achieve competitive control performance across diverse POMDPs, while using less than half the number of trainable parameters required by their fully trainable counterparts.
Continual post-training faces a dual challenge of acquisition and retention: models must learn new capabilities while preserving prior knowledge. While Self-Distillation Fine-Tuning (SDFT) mitigates this using self-supervision with in-context demonstrations, it relies on strong in-context learning and instruction-follo...
Woosung Koh, Hyunsoo Lee, Kyungjae Lee et al.· 0 citations
The optimization objective for e-commerce advertising creatives has shifted from aesthetic quality to online business metrics like click-through rate (CTR). Recent methods generate appealing ad images from product images by constructing CTR-based reward signals and refining prompt generation via Reinforcement Learning....
Xu Zhang, Haohan Wang, Zheng Zhang et al.· 0 citations
Counterfactual text editing is the ability to revise a given text in light of a hypothetical change while preserving facts not causally affected by that change. When asked what would have happened if a doctor had not recommended surgery, the model output should retain the patient's medical history and modify only the c...
Arvind Raghavan, Inwoo Hwang, Shreyas Havaldar et al.· 0 citations
Multi-view representations enable pretrained image and video vision-language models to understand 3D scenes, but their long visual token sequences make inference costly. Redundancy arises both from repeated observations of the same scene regions across views and from processing all visual tokens at every LLM layer, reg...
Yong-Xing Zheng, Hao Wu, Fang Li et al.· 0 citations
As LLMs become increasingly capable of implementing software, a critical bottleneck emerges: whether they truly understand what users want. Users often begin with vague intent, leaving important requirements to be clarified through interaction. In repository-level development, misunderstandings can propagate and cause...
Ming-Yu Chen, Zhiqing Zhong, Man-Yi Wang et al.· 0 citations
Reinforcement learning has become the foundational paradigm for enhancing the reasoning capabilities of LLMs. Yet the dominant approach, Proximal Policy Optimization (PPO), relies on a geometrically degenerate clipping proxy that cannot adapt its resistance to policy shifts, leading to premature clipping and degraded e...
Zhi-Cheng Cai, Xin-Yuan Guo, Han Wu et al.· 0 citations
