Industrial recommendation rankers keep scaling in width and depth to billions of parameters, but this increasingly incurs costs in parameter growth, distributed communication overhead, and training inefficiency. We introduce \underline{C}hain \underline{D}imension \underline{S}caling (CDS), a scaling strategy orthogona...
Zhimeng Jiang, Yuening Li, Ruixiao Sun et al.· 0 citations
As IT systems grow in complexity, automated agent-driven incident response becomes increasingly critical. A key capability for these agents is summarizing large log streams into human-readable reports. Further progress requires meaningfully evaluating this function, yet existing benchmarks leave end-to-end summarizatio...
Valerii Iakovlev, Markus Spanring, Martin Flechl· 0 citations
Web application development often proceeds through successive requests that extend an existing implementation. Yet web-generation benchmarks primarily evaluate development toward fixed specifications. We introduce EvoWeb, a novel benchmark comprising 320 cumulative requests over 32 web applications, grounded in real-wo...
David Nukrai, Bar Alon, Nativ Levy et al.· 2 citations
The framework, Recursive Environment Scaling (RES), transforms existing benchmark environments while preserving their objectives and completion requirements and identifies recursive inheritance and dependency composition as important contributors to increased difficulty.
Lin Jing, Zheng-Bo Jiao, Yi-Chen Xiang et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Multi-step retrosynthesis planning is often considered solved once a route to purchasable starting materials is found, but such a route should also be practical in real-world chemistry. Recent LLM-based planners formulate this process as sequential decision-making over an AND-OR graph, yet typically expand a single mol...
Yejun Kim, Yongjun Jeong, Seongjun Yun et al.· 0 citations
Video world models must generate physically plausible futures, not just realistic frames. Existing approaches typically use semantic, physical, or predictive information to guide the visual generation stream, leaving a fundamental question: where inside the generator should predictions of scene evolution live? Standard...
Shinseong Kim, Jaeseok Jeong, Jianlong Fu et al.· 0 citations
The results support ChronoScribe's division of labor, in which the LLM handles interpretation, configuration, and communication while the TSFM handles numerical forecasting.
Yu-Chen Jiao, Miao Zhang, Yuxuan Liu et al.· 0 citations
Multi-agent debate has shown promise for improving the reasoning of large language models, yet recent theory suggests its benefits are highly regime-dependent. While interaction can amplify informative signals under corrective conditions, symmetric debate dynamics are neutral in expectation, often making majority votin...
Seungdong Yoa, Ye Seul Sim, Suhee Yoon et al.· 0 citations
As AI agents increasingly act through external tools, tool results are no longer simple outputs but execution artifacts that must remain correctly associated with the context that produced them. Yet as results persist across caches, recovery mechanisms, and shared runtime state, this association can break, causing a re...
Yu-Tong Zhi, Jie Zhang, Xiu-Qi Lu et al.· 0 citations
When a steering intervention changes a language model's behavior, which part of the intervention deserves the credit? We test this question for a target-centered, scaled conceptor complement by removing its fitted covariance filter while preserving the same target-specific center and gain. At the original aperture $\al...
HyWe, a framework that unifies efficient information gathering and browser interaction in a single 35B web agent, is introduced, and a fully automatic pipeline that uses task prototypes, four switching patterns, and multi-agent synthesis to produce verified hybrid trajectories is developed.
Zi-Jian Cai, Hao-Wei Liu, Kai-Lai Chen et al.· 0 citations
Test-time agentic computation can improve multimodal large language models (MLLMs) through planning, tool use, iterative perception, and reflection, but indiscriminate computation may duplicate resolved work, introduce noisy evidence, and perturb correct predictions. We propose that such computation should be residual:...
Yi-Ke Wu, Keyan Chen, Qi-Xiong Wang et al.· 0 citations
