ReForge:用经验证的 LLM 修订让 ABR 算法持续进化
ReForge: Keeping ABR Algorithms Never Finished with Verified Large Language Model Revisions
AI 导读
arXiv 论文提出 ReForge 框架,让 LLM 在 ABR 算法部署后持续修订码率选择规则,且不损害已有网络的性能。系统保留既有 ABR 策略池不变,仅当 GPU 模拟器回放验证修订能整体提升且不使任何已服务网络变差时才部署。
正文 · 原文
Abstract:Designing an ABR algorithm for a network scenario takes an engineer months, and large language models (LLMs) now do this work in hours. But either way, the algorithm is fixed once it is deployed. It fits the networks seen during its design and falls behind on the networks that arrive afterwards. We ask whether an ABR algorithm can keep being designed after deployment, revised as each new network arrives, without becoming worse on any network it already serves. We propose ReForge, which keeps a pool of existing ABR policies unchanged and lets an LLM revise only the rules that decide which policy chooses the next bitrate. When a new network arrives, the LLM reads how the current rules and every policy perform on the networks served so far and rewrites the rules. The rewrite is deployed only if a replay in simulation shows that it improves the served networks as a whole and makes none of them worse. A fast GPU simulator that replays every episode exactly makes this acceptance test cheap enough to run for every revision. We evaluate ReForge on nine real-world network datasets, with new networks arriving one at a time. The revisions raise the mean QoE from 1.43 to 1.79, past the best single policy at 1.66 and close to the 1.86 of selecting the best policy for each dataset, while the QoE on the networks served first does not drop. The gain also holds in a real video player. All code, data, and records will be available.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2608.15138 [cs.AI] |
| (or arXiv:2608.15138v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2608.15138 arXiv-issued DOI via DataCite |
Submission history
From: Zhiqiang He [view email]
[v1]
Sat, 15 Aug 2026 09:23:24 UTC (614 KB)
[v2]
Fri, 13 Nov 2026 05:22:37 UTC (932 KB)
来源:arXiv cs.AI · arxiv.org