{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:ISSIX3ZZF7LWCM7INRUGIBQEA5","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":"8e032954d882f6fc272004c942c40ae072f6638fde32ea039a9ae82a692907fd","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-15T04:54:08Z","title_canon_sha256":"b8d22527de8290f1cbf0c5e51d307dfc379eb50737ec5e89cbba61895e58a1ea"},"schema_version":"1.0","source":{"id":"2605.15615","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.15615","created_at":"2026-05-20T00:01:08Z"},{"alias_kind":"arxiv_version","alias_value":"2605.15615v1","created_at":"2026-05-20T00:01:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.15615","created_at":"2026-05-20T00:01:08Z"},{"alias_kind":"pith_short_12","alias_value":"ISSIX3ZZF7LW","created_at":"2026-05-20T00:01:08Z"},{"alias_kind":"pith_short_16","alias_value":"ISSIX3ZZF7LWCM7I","created_at":"2026-05-20T00:01:08Z"},{"alias_kind":"pith_short_8","alias_value":"ISSIX3ZZ","created_at":"2026-05-20T00:01:08Z"}],"graph_snapshots":[{"event_id":"sha256:632497433ed21f14f73225aeac7d1baec8e2b22117650b4d890476bd095e7f27","target":"graph","created_at":"2026-05-20T00:01:08Z","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":[{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T19:34:34.617498Z","status":"skipped","version":"1.0.0"},{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-19T17:41:56.042299Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.15615/integrity.json","findings":[],"snapshot_sha256":"53c3ac7f48028fefe1c5f10cc057a117803e3e6dbad5cb3d43ad5d6d6bdaf5f1","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Efficient transfer learning of vision-language models (VLMs) commonly suffers from a Base-New Trade-off (BNT): improving performance on unseen (new) classes often degrades accuracy on known (base) classes. Addressing how to boost recognition of unseen classes without sacrificing known-class performance remains a central challenge. Existing work often simplistically attributes the BNT to overfitting on known classes. We observe an interesting phenomenon: VLMs frequently exhibit asymmetric confusion on certain downstream data, i.e., samples of class A are systematically mispredicted as class B, ","authors_text":"Senmao Tian, Shunli Zhang, Xiang Wei","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-15T04:54:08Z","title":"Neutral-Reference Prompting for Vision-Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.15615","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":"sha256:6bb8a986bf92783f081f634791e734d18fae5fa24c4fa4a3f15111b324ef06a4","target":"record","created_at":"2026-05-20T00:01:08Z","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":"8e032954d882f6fc272004c942c40ae072f6638fde32ea039a9ae82a692907fd","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-15T04:54:08Z","title_canon_sha256":"b8d22527de8290f1cbf0c5e51d307dfc379eb50737ec5e89cbba61895e58a1ea"},"schema_version":"1.0","source":{"id":"2605.15615","kind":"arxiv","version":1}},"canonical_sha256":"44a48bef392fd76133e86c686406040750bf09f410562d3bbe4e587a64844d3c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"44a48bef392fd76133e86c686406040750bf09f410562d3bbe4e587a64844d3c","first_computed_at":"2026-05-20T00:01:08.257119Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:01:08.257119Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"pbInBPwldOtHFV2JigvlDvS1Zt1DnEK81VmFu0VwntowfLrSQv11h+DaMHRaKku3lCrAnsp35Q3LsVT2iV0wDg==","signature_status":"signed_v1","signed_at":"2026-05-20T00:01:08.258020Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.15615","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6bb8a986bf92783f081f634791e734d18fae5fa24c4fa4a3f15111b324ef06a4","sha256:632497433ed21f14f73225aeac7d1baec8e2b22117650b4d890476bd095e7f27"],"state_sha256":"174392c37638a79af61c0bb4fc04dfed27599fea5aab268f5a2249fa71539b94"}