Feature caching has become a mainstream training-free strategy for accelerating video diffusion models by reusing intermediate outputs across denoising steps. However, existing adaptive methods rely on heuristic difference measures misaligned with actual generation quality, employ a single global threshold that ignores...
Ji-Yue Jiang, Zhi-Yu Huo, Yu-Han Chen et al.· 0 citations
We present PatTransBench, a 12,000-item Chinese-English/Japanese/Korean patent machine translation (PatMT) benchmark with a four-dimensional evaluation protocol. The benchmark comprises 6,000 short-text (sentence/claim) and 6,000 long-text (paragraph) pairs, evenly split across three language pairs and both translation...
Developing reliable predictive models for multi-physics systems often forces a compromise between computationally demanding physical simulators and data-hungry, opaque machine learning (ML) algorithms. While incorporating dimensional analysis into directed computation graphs offers a structurally sound alternative to e...
Akshay Dhalpe, Arnab Chattopadhyay, S. B. Balani et al.· 0 citations
Generative Social Simulation (GSS) leverages Large Language Models (LLMs) to create sophisticated personas that define agents with demographics, emotions, and political opinions. While prior work shows that personas compress outputs into stereotypical caricatures, there is no comprehensive study of how these caricature...
Dimitris Michailidis, Petter Törnberg, Fernando P. Santos· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Speculative Recursive Self-Improvement (SpecRSI) is introduced, which drafts executable successor agents before their predecessors are fully evaluated, while requiring complete verification before adoption.
G GrowRL is proposed, an RL method that organizes learned updates into compact computational pathways through structural growth, and shows competitive mathematical reasoning and out-of-distribution performance with a small number of trainable parameters.
Mixture-of-Experts (MoE) models scale model capacity efficiently by activating only a small subset of experts for each input, while still requiring all parameters to be stored in memory, resulting in substantial memory overhead. Expert merging techniques have recently emerged as a promising compression paradigm that re...
Dohyeon Kim, Bedionita Soro, Bruno Andreis et al.· 0 citations
We present PatMT2 (Patent Machine Translation, v2), a two-stage LoRA pipeline that adapts the Hy-MT2-7B dedicated translation model to four-language (Chinese/English/Japanese/Korean) patent translation. On six translation directions, the Stage-2 checkpoint reaches a macro-average BLEU of 43.39 (BF16) to 43.57 (AWQ INT4...
Fine-tuning large language models at scale is increasingly limited by the memory cost of backpropagation. Zeroth-Order (ZO) optimizers reduce this burden by relying only on forward function evaluations, but they still exhibit a substantial performance gap to first-order (FO) optimizers such as Adam. We argue that this...
This work proposes ChainSleuth, a graph-tool-augmented agent that iteratively reasons, invokes graph tools, and updates its tracing trajectory, enabling adaptive information acquisition and recovery from erroneous paths in multi-hop fund tracing.
Ru-Nang He, Tong-Ya Zheng, Jing-Yao Cao et al.· 0 citations
Current image-text evaluation often relies on a global score, which gives limited insight into which semantic details are supported by the image. Explainable metrics provide rationales, but the visual evidence supporting those rationales can still be difficult to verify. We introduce VisExMEM, a reference-free evaluato...
Subhadeep Roy, Christoph Leiter, Steffen Eger· 0 citations
Medication recommendation requires jointly considering diagnoses, comorbidities, longitudinal patient history, and treatment interactions. Although recent retrieval-augmented generation approaches using language models have shown promise for clinical recommendation, most existing systems rely on fixed retrieval pipelin...
Kyungkeon Chung, Se Un Park, Hyejin Lee et al.· 0 citations
