{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:7C53S2RCYVBEV4RJMXKOND544V","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"768f9ab4abfd6b010062b86b2627885d16de488cf9ed69fe6f56eba1dc5361fa","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-23T12:50:18Z","title_canon_sha256":"adcf42aa0868c43d10cda759c147949cf02473513533091f7ab3059533fae88e"},"schema_version":"1.0","source":{"id":"2607.21273","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.21273","created_at":"2026-07-24T01:24:23Z"},{"alias_kind":"arxiv_version","alias_value":"2607.21273v1","created_at":"2026-07-24T01:24:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.21273","created_at":"2026-07-24T01:24:23Z"},{"alias_kind":"pith_short_12","alias_value":"7C53S2RCYVBE","created_at":"2026-07-24T01:24:23Z"},{"alias_kind":"pith_short_16","alias_value":"7C53S2RCYVBEV4RJ","created_at":"2026-07-24T01:24:23Z"},{"alias_kind":"pith_short_8","alias_value":"7C53S2RC","created_at":"2026-07-24T01:24:23Z"}],"graph_snapshots":[{"event_id":"sha256:d109b3241c2dae5cbf3f8d925b370002755c00b1cc160abf42b1dfb8820ffad6","target":"graph","created_at":"2026-07-24T01:24:23Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2607.21273/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Dense per-step supervision is an appealing remedy for sparse-reward, long-horizon LLM agents: reward the agent for predicting its next observation, and memory should follow. We show that under group-normalized RL (GRPO), this recipe does not merely fail -- it destroys the policy. Across Qwen3-1.7B/4B/8B on ALFWorld, a potential-based prediction reward drives every run into a degenerate absorbing state (prediction accuracy -> 1.0, task success -> 0,episode length pinned at the horizon): the \"dark room\" pathology, built automatically by the optimizer. A single-factor ablation localizes the cause","authors_text":"Yu Wang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-23T12:50:18Z","title":"The Dark Room in the Reward Channel: Dense Prediction Rewards Collapse GRPO-Trained LLM Agents -- and What Actually Works"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.21273","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:d46886843b45a494af80847a6f76bdbc9376a48a117e70cd5e9bf532917f2a58","target":"record","created_at":"2026-07-24T01:24:23Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"768f9ab4abfd6b010062b86b2627885d16de488cf9ed69fe6f56eba1dc5361fa","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-23T12:50:18Z","title_canon_sha256":"adcf42aa0868c43d10cda759c147949cf02473513533091f7ab3059533fae88e"},"schema_version":"1.0","source":{"id":"2607.21273","kind":"arxiv","version":1}},"canonical_sha256":"f8bbb96a22c5424af22965d4e68fbce568cf9d1bfcea6f2520552a063ef2a088","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f8bbb96a22c5424af22965d4e68fbce568cf9d1bfcea6f2520552a063ef2a088","first_computed_at":"2026-07-24T01:24:23.132098Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-24T01:24:23.132098Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7HUCiJ4AotF9gI47j6ATEAdG8XlGxuwRLGh+TWaNgS/j2IfsR8eWW7ZK+Oif4r1v6hZt/EI5FPS6br2G45AoAw==","signature_status":"signed_v1","signed_at":"2026-07-24T01:24:23.133052Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.21273","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d46886843b45a494af80847a6f76bdbc9376a48a117e70cd5e9bf532917f2a58","sha256:d109b3241c2dae5cbf3f8d925b370002755c00b1cc160abf42b1dfb8820ffad6"],"state_sha256":"d75d202cd20376ce0ead5a69287b9d01666569fe2a641a96f72ddb4916e734ed"}