Abstract
We present onPanda, an interactive tool for efficiently annotating LLM alignment data and agent trajectories. onPanda adopts token-level correction as its core interaction: while reading a model response, the annotator locates the first inappropriate token and either picks a substitute from the model's candidate tokens or types the correct text via free-form editing. The system then truncates everything after that position and continues generation from the corrected prefix, repeating this locate-correct-continue loop until a satisfactory response is obtained. This mechanism lets annotators precisely steer model outputs at low cost: a small controlled study suggests that onPanda reduces median annotation time by 52% over manual post-editing. Since the vast majority of tokens in the final response are generated by the model itself, the resulting data largely preserves the model's sampling distribution and is well suited for constructing on-policy SFT and preference data. Furthermore, the token-level corrections recorded during annotation provide fine-grained supervision with precise positions and naturally paired positive--negative samples. onPanda also connects to external tools and harnesses, enabling interactive trajectory annotation in realistic environments. In addition, we release Panda-CVL, a dataset annotated with onPanda, together with a benchmark for token-level correction.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Yang, L., Liu, M., Wang, J., Guo, H., Zhao, L., Ge, Z., An, K., Jiao, B., Han, Q., Jiang, D., Shen, S., & Zhang, X. (2026). onPanda: Efficient Annotation of On-Policy Alignment Data for LLMs and Agents via Token-Level Correction. https://omanscience.com/en/articles/onpanda-efficient-annotation-of-on-policy-alignment-data-for-llms-and-agents-via-token-level-correction
MLA 9
Yang, Lei, et al. "onPanda: Efficient Annotation of On-Policy Alignment Data for LLMs and Agents via Token-Level Correction." https://omanscience.com/en/articles/onpanda-efficient-annotation-of-on-policy-alignment-data-for-llms-and-agents-via-token-level-correction.
Chicago (author–date)
Yang, Lei, Mengyin Liu, Jia Wang, Hangyu Guo, Liang Zhao, Zheng Ge, Kang An, Binxing Jiao, Qi Han, Daxin Jiang, Siqi Shen, and Xiangyu Zhang. 2026. "onPanda: Efficient Annotation of On-Policy Alignment Data for LLMs and Agents via Token-Level Correction." https://omanscience.com/en/articles/onpanda-efficient-annotation-of-on-policy-alignment-data-for-llms-and-agents-via-token-level-correction.
Harvard
Yang, L., Liu, M., Wang, J., Guo, H., Zhao, L., Ge, Z., An, K., Jiao, B., Han, Q., Jiang, D., Shen, S. and Zhang, X. (2026) 'onPanda: Efficient Annotation of On-Policy Alignment Data for LLMs and Agents via Token-Level Correction', Available at: https://omanscience.com/en/articles/onpanda-efficient-annotation-of-on-policy-alignment-data-for-llms-and-agents-via-token-level-correction.
Vancouver
Yang L, Liu M, Wang J, Guo H, Zhao L, Ge Z, et al. onPanda: Efficient Annotation of On-Policy Alignment Data for LLMs and Agents via Token-Level Correction. https://omanscience.com/en/articles/onpanda-efficient-annotation-of-on-policy-alignment-data-for-llms-and-agents-via-token-level-correction
IEEE
L. Yang, M. Liu, J. Wang, H. Guo, L. Zhao, Z. Ge, K. An, B. Jiao, Q. Han, D. Jiang, S. Shen, and X. Zhang, "onPanda: Efficient Annotation of On-Policy Alignment Data for LLMs and Agents via Token-Level Correction," https://omanscience.com/en/articles/onpanda-efficient-annotation-of-on-policy-alignment-data-for-llms-and-agents-via-token-level-correction.