Authors

Tianqi Bu

Publications 2

Preprint Open access

Re-derivability Decides What a Staged Agent Pipeline Recovers After an Upstream Fault

Tianqi Bu, YuXuan Peng, Junteng Tu et al. · 2026

One variable sets what an upstream fault costs a staged pipeline of language-model agents: re-derivability, how much of what a stage needs it can rebuild from the original problem. Grounding an inspector agent in that problem is worth +0.608 [+0.517, +0.700] to +0.358 over a blind one on four open-weight backbones serv …

Preprint Open access

The Decomposition Tax: LLM Pipelines Lose Up to 40 Accuracy Points at Their Own Interfaces

Tianqi Bu, YuXuan Peng, Junteng Tu et al. · 2026

A four-stage LLM pipeline gives up as much as 40.5 accuracy points at its own interfaces (gemma-3-12B on MATH-500, Holm-corrected p = 1.66e-19; the largest tax in the primary family). We hold model, problem, stages, stage prompts and completion budget fixed, vary only whether each stage can still see the original probl …

Co-authors

YuXuan Peng Junteng Tu Henghui Xiao