mlsanityData to 14 Oct 2026

Pangu-ACE: Adaptive Cascaded Experts for Educational Response Generation on EduBench

arXiv 2604.14828 ↗First posted 16 Apr 2026cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

Educational assistants should spend more computation only when the task needs it. This paper rewrites our earlier draft around the system that was actually implemented and archived in the repository: a sample-level 1B to 7B cascade for the shared-8 EduBench benchmark. The final system, Pangu-ACE, uses a 1B tutor-router to produce a draft answer plus routing signals, then either accepts the draft or escalates the sample to a 7B specialist prompt. We also correct a major offline evaluation bug: earlier summaries over-credited some open-form outputs that only satisfied superficial format checks. After CPU-side rescoring from saved prediction JSONL, the full Chinese test archive (7013 samples) shows that cascade_final improves deterministic quality from 0.457 to 0.538 and format validity from 0.707 to 0.866 over the legacy rule_v2 system while accepting 19.7% of requests directly at 1B. Routing is strongly task dependent: IP is accepted by 1B 78.0% of the time, while QG and EC still escalate almost always. The current archived deployment does not yet show latency gains, so the defensible efficiency story is routing selectivity rather than wall-clock speedup. We also package a reproducible artifact-first paper workflow and clarify the remaining external-baseline gap: GPT-5.4 re-judging is implemented locally, but the configured provider endpoint and key are invalid, so final sampled-baseline alignment with GPT-5.4 remains pending infrastructure repair.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 18 of its 21 references that are in the map. First posted on arXiv 16 Apr 2026.

Builds on

18 of 21 references.

References · 18

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  2. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  3. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  4. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,708
  5. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,276
  6. 2023Toolformer: Language Models Can Teach Themselves to Use Toolscited by 4,182
  7. 2022FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awarenesscited by 3,673
  8. 2023FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioningcited by 2,598
  9. 2023A Survey on Large Language Model based Autonomous Agentscited by 2,242
  10. 2022Fast Inference from Transformers via Speculative Decodingcited by 1,653
  11. 2023FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performancecited by 853
  12. 2024Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Headscited by 843
6 more
  1. 2024RouteLLM: Learning to Route LLMs with Preference Datacited by 682
  2. 2024EAGLE: Speculative Sampling Requires Rethinking Feature Uncertaintycited by 578
  3. 2024Quiet-STaR: Language Models Can Teach Themselves to Think Before Speakingcited by 301
  4. 2024Large Language Models for Education: A Survey and Outlookcited by 188
  5. 2023System 2 Attention (is something you might need too)cited by 89
  6. 2025Pangu Embedded: An Efficient Dual-system LLM Reasoner with Metacognitioncited by 62

Built on by

Not cited yet.