Chinese AI lab Zhipu AI releases GLM-5.2 with a stable 1-million-token context under the MIT license. On hours-long coding tasks, the open-source model trails Anthropic's Opus models by just a few ...

description [ACL 2026 Findings][LLM推理][多教师蒸馏] 作者提出 CoRD（Collaborative Reasoning Decoding），把多教师 Long-CoT 推理蒸馏从「先生成完整轨迹再后选」改造成「step-wise 协同解码」——每步让多个 LRM 提议候选 step，用 meta-prover 的 predictive perplexity ...

Abstract: Glaucoma is a leading cause of irreversible blindness worldwide, with asymptomatic early stages often delaying diagnosis and treatment. Early and accurate diagnosis requires integrating ...

The Information Networking Institute (INI) offers students the flexibility to explore unique topics through a master’s thesis, development-based project or area of concentration. These options are ...

Official implementation of SeerAttention and SeerAttention-R - a novel trainable sparse attention mechanism that learns intrinsic sparsity patterns directly from LLMs through self-distillation at post ...

Abstract: Semantic communication (SC) offers promising advancements in data transmission efficiency and reliability by focusing on delivering true meaning rather than solely binary bits of messages.

In the field of structured information extraction, there are typically semantic and syntactic constraints on the output of information extraction (IE) systems. These constraints, however, can ...