{"ID":2923581,"CreatedAt":"2026-06-02T04:05:25.881865328Z","UpdatedAt":"2026-06-04T13:12:39.622923895Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2606.02386","arxiv_id":"2606.02386","title":"AgentPLM: Agentic Protein Language Models with Reasoning-Augmented Decoding for Protein Sequence Design","abstract":"Protein language models (PLMs) are passive oracles: they generate sequences in a single forward pass with no mechanism to consult external biophysical feedback or redirect generation when a candidate violates thermodynamic or structural constraints. We introduce AgentPLM, which addresses this by equipping a pre-trained PLM with i) Reasoning-Augmented Decoding (RAD), which interleaves autoregressive generation with tool calls (ESMFold, FoldX, AutoDock Vina), and ii) Contrastive Agent Policy Optimisation (CAPO), a trajectory-level extension of direct preference optimisation that trains the policy end-to-end to learn when oracle feedback is informative rather than merely imitating high-fitness sequences. We evaluate AgentPLM on benchmark tasks spanning de novo enzyme design, antibody optimisation, thermostability, PPI interface design, and zero-shot fitness prediction with standardised oracle APIs and controlled sequence-identity splits. AgentPLM achieves state-of-the-art results with a gain in antibody top-10% hit rate over the strongest passive baseline, providing mechanistic evidence of online error correction without explicit backtracking.","short_abstract":"Protein language models (PLMs) are passive oracles: they generate sequences in a single forward pass with no mechanism to consult external biophysical feedback or redirect generation when a candidate violates thermodynamic or structural constraints. We introduce AgentPLM, which addresses this by equipping a pre-trained...","url_abs":"https://arxiv.org/abs/2606.02386","url_pdf":"https://arxiv.org/pdf/2606.02386v1","authors":"[\"Sahil Rahman\",\"Maxx Richard Rahman\"]","published":"2026-06-01T15:35:02Z","proceeding":"cs.AI","tasks":"[\"cs.AI\",\"q-bio.QM\"]","methods":"[\"Language Model\"]","has_code":false}