dair-ai 一周论文精选：HySparse2 与 Harness-Zero 等 7 篇入选

Two-level KV sharing targets the long-context prefill bottleneck that agentic workloads hit, cutting compute without a full attention redesign. Engineers building long-context inference stacks can benchmark the approach against hybrid SWA baselines on similar MoE setups. The MiMo-V3 linkage also signals how Xiaomi positions its model stack against domestic rivals like DeepSeek and Qwen.

Primary sourcepandaily@[email protected] (Pandaily)Source

Evaluation shifts from comparing static code screenshots to judging how agents actually behave inside an app, which changes how generation quality can be measured for web development agents. Builders of coding agents and eval pipelines can use the two open-sourced trace datasets to train reward models or benchmark correctness without collecting their own user complaints data. Compared with human-preference-only arenas, this adds an automated signal that scales with computer-use capability.

lmarena.ai@lmarena_aiSource

The core takeaway is that evaluation signal can be harvested from real user interactions instead of hand-written test suites, shifting agent benchmarking toward live platform data. Researchers and teams building computer-use agents can apply this pattern to score traces with LLM judges or reward models trained on user complaints. Code Arena and Agent Arena already supply the battle and complaint data, giving practitioners a concrete starting point for evaluation pipelines.

lmarena.ai@lmarena_aiSource