Selected work

Representative Papers

RRC: Unlocking Generative Reward Models in LLM Reinforcement Learning via Ranking-Based Reward Construction

Aug 06, 2026

This work addresses the challenge that generative reward models, due to their comparative outputs, are incompatible with the scalar rewards required by reinforcement learning, thereby hindering effective training of large language models. To overcome this limitation, the paper proposes a Ranking-based Reward Construction (RRC) method, which innovatively introduces self-competitive ranking and anchor-guided ranking strategies to transform relative preference rankings into scalar reward signals suitable for reinforcement learning. By circumventing the constraints of conventional scalar reward formulation, RRC achieves substantially improved training performance on open-ended dialogue and reasoning benchmarks, consistently outperforming existing reward modeling approaches.

0 citationsRead paper

RusLICA: A Russian-Language Platform for Automated Linguistic Inquiry and Category Analysis

Jan 28, 2026

This study addresses the lack of psycholinguistic text analysis tools tailored to the grammatical and cultural specifics of Russian. To bridge this gap, the authors present RusLICA, the first automated platform for Russian psycholinguistic analysis, which avoids direct translation of existing dictionaries and instead constructs an original psycholinguistic taxonomy grounded in native Russian corpora and semantic resources. Integrating lemmatization, multi-source semantic lexicons, corpus-based methods, and pretrained language models, RusLICA extracts 96-dimensional textual features and successfully maps lexical items to 42 psycholinguistic categories. The platform enables fine-grained quantification of linguistic and psychological characteristics in Russian texts and is publicly available online, thereby filling a critical void in computational psycholinguistics for the Russian language.

0 citationsRead paper

Recent publications

Latest Papers

RRC: Unlocking Generative Reward Models in LLM Reinforcement Learning via Ranking-Based Reward Construction

Aug 06, 2026

This work addresses the challenge that generative reward models, due to their comparative outputs, are incompatible with the scalar rewards required by reinforcement learning, thereby hindering effective training of large language models. To overcome this limitation, the paper proposes a Ranking-based Reward Construction (RRC) method, which innovatively introduces self-competitive ranking and anchor-guided ranking strategies to transform relative preference rankings into scalar reward signals suitable for reinforcement learning. By circumventing the constraints of conventional scalar reward formulation, RRC achieves substantially improved training performance on open-ended dialogue and reasoning benchmarks, consistently outperforming existing reward modeling approaches.

0 citationsRead paper

RusLICA: A Russian-Language Platform for Automated Linguistic Inquiry and Category Analysis

Jan 28, 2026

This study addresses the lack of psycholinguistic text analysis tools tailored to the grammatical and cultural specifics of Russian. To bridge this gap, the authors present RusLICA, the first automated platform for Russian psycholinguistic analysis, which avoids direct translation of existing dictionaries and instead constructs an original psycholinguistic taxonomy grounded in native Russian corpora and semantic resources. Integrating lemmatization, multi-source semantic lexicons, corpus-based methods, and pretrained language models, RusLICA extracts 96-dimensional textual features and successfully maps lexical items to 42 psycholinguistic categories. The platform enables fine-grained quantification of linguistic and psychological characteristics in Russian texts and is publicly available online, thereby filling a critical void in computational psycholinguistics for the Russian language.

0 citationsRead paper