AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

TraceRouter++: Causal Semantic Routing Realignment via Optimal Transport

Chuancheng Shi, Shangze Li, Wenhua Wu, Hengzhi Chen, Fei Shen, Zhuo Li, Zhiyong Wang, Tat‐Seng Chua

TPAMI 2026 · Large language models › Language, safety and interpretability · approach: Safety alignment of LLMs · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add TraceRouter++ routing realignment to LLaMA3-8B-InstructLLaMA3-8B-Instruct · LLM jailbreak suiteavg DSR: 63.2 -> 99.5Improved
add TraceRouter++ to Mistral-7B-InstructMistral-7B-Instruct · LLM jailbreak suiteavg DSR: 6.6 -> 98.8Improved
upgrade TraceRouter to TraceRouter++TraceRouter · LLaMA3-8B jailbreak suiteavg DSR: 99.2 -> 99.5Improved
replace linear steering with OT realignment (ablation)TraceRouter++ · Stable Diffusion 1.4DSR: 72.2 -> 99.3Improved
replace isolated neuron suppression with continuous path realignment (ablation)TraceRouter++ · LLaVA-1.5-7BDSR: 77.4 -> 99.4Improved
replace isolated neuron suppression with path realignment (ablation)TraceRouter++ · Stable Diffusion 1.4FID: 19.12 -> 8.62Improved
replace linear steering with OT realignment (ablation)TraceRouter++ · LLaVA-1.5-7BMMBench: 45.6 -> 47.4No clear effect
use SAE expansion 4x instead of 2x (ablation)TraceRouter++ · LLaMA3-8B-InstructDSR: 88.4 -> 99.5Improved
use SAE expansion 4x instead of 8x (ablation)TraceRouter++ · LLaVA-1.5-7BDSR: 76.4 -> 99.4Improved
increase OT intensity lambda from 1 to 3 (ablation)TraceRouter++ · LLaMA3-8B-InstructORR: 3.5 -> 12.4Hurt
increase OT intensity lambda from 1 to 3 (ablation)TraceRouter++ · LLaMA3-8B-InstructGlobal-MMLU-Lite: 71.1 -> 68.5Hurt
increase OT intensity lambda from 0.5 to 1 (ablation)TraceRouter++ · LLaMA3-8B-Instructavg DSR: 82.8 -> 99.5Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy