Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Topology-Enhanced Alignment for Large Language Models: Trajectory Topology Loss and Topological Preference Optimization

About

Alignment of large language models (LLMs) via SFT and RLHF/DPO typically ignores the global geometry of the representation space, relying instead on local token likelihoods or scalar scores. We view generation as tracing a semantic trajectory in hidden space and propose a topology-enhanced alignment framework that regularizes these trajectories using 0-dimensional persistent homology. First, for SFT, we introduce Trajectory Topology Loss (TTL). Treating prompt and gold-answer embeddings as a mixed point cloud, we use a 0D persistent homology algorithm to extract "prompt-answer bridges." TTL aligns the model's actual update direction with these topological bridges rather than arbitrary directions. Second, for DPO, we propose Topological Preference Optimization (TPO). TPO constructs topic-specific semantic preference vectors and aligns the improvement direction between rejected and chosen responses with these vectors in an intermediate hidden layer. We also introduce a dynamic weighting scheme to balance DPO and TPO losses. Evaluating on Qwen2.5-7B-Instruct using UltraChat and Anthropic HH-RLHF, our topology-enhanced objectives consistently outperform strong non-topological baselines (e.g., per-example, nearest-neighbor, random regularizers) on automatic preference metrics and LLM-judge evaluations, while maintaining or improving toxicity. Results show persistent homology and trajectory geometry offer a promising direction for controllable alignment.

Yurui Pan, Ke Xu, Bo Peng• 2026

Related benchmarks

TaskDatasetResultRank
Instruction FollowingIFEval
IFEval Accuracy71.8
836
Instruction FollowingAlpacaEval 2.0
Win Rate55.6
722
Multi-turn Dialogue EvaluationMT-Bench
Overall Score8.88
532
Instruction FollowingAlpacaEval
Win Rate56.5
420
Reward ModelingRewardBench--
216
Multi-turn dialogueMT-Bench
MT-Bench Score8.81
126
Harmlessness evaluationHH-RLHF
Harmlessness Rate94.5
6
Reward Modeling EvaluationRewardBench
R-Bench Score88.1
3
Alignment EvaluationHH-RLHF (test)
Reward Model Score65.4
2
Instruction FollowingUltraChat
RM Score67.8
2
Showing 10 of 10 rows

Other info

Follow for update