{"_canonicalization":{"envelope_id":"axm_ + sha256(envelope minus {signature, axiom_id, anchors})","envelope_signature":"ed25519(envelope minus {signature, axiom_id})","json":"sort_keys=True, separators=(',',':'), ensure_ascii=False, allow_nan=False, utf-8","leaf_hash":"sha256(0x00 || canonical_json(envelope_full))","seal_signature":"ed25519(seal minus {signature, sig_algorithm})"},"axiom_id":"axm_2a61018a51df6a2ee5453ae39a1ad78acf8e622d7875d83f88743d9c9b043b5c","bitcoin_anchor":{"bitcoin_attestations":[],"calendar_attestations":[],"ots_url":"","stamped_at":"","status":"pending_next_stamp"},"envelope":{"anchors":[{"chain":"crovia.axiom_graph","height":0,"merkle_proof":"spider_vendor_press_v1","root_at_anchor":"spider_vendor_press_v1"}],"axiom_id":"axm_2a61018a51df6a2ee5453ae39a1ad78acf8e622d7875d83f88743d9c9b043b5c","axiom_type":"AX.OBS","body":{"axiom_subtype":"news.vendor_press.v1","category":"news","fingerprint":"f6d5e46c1a52e857b323f4126341b6612eb8913b1c25b8bb063bd7021785330f","published":"Fri, 10 Jul 2026 00:00:00 -0400","receipt_hash":"f6d5e46c1a52e857b323f4126341b6612eb8913b1c25b8bb063bd7021785330f","schema":"spider.news.vendor_press.v1","spider":"vendor_press","spider_record":{"axiom_subtype":"news.vendor_press.v1","category":"news","decision_hint":"POSITIVE","envelope_target":"AX.OBS","fingerprint":"f6d5e46c1a52e857b323f4126341b6612eb8913b1c25b8bb063bd7021785330f","observed_at":"2026-07-10T04:43:53.465232Z","parent_run_hash":"06997be187ba20932a2030c56de194579eacf484085252a25bee544eab183e91","published":"Fri, 10 Jul 2026 00:00:00 -0400","runtime_version":"0.1.0","schema":"spider.news.vendor_press.v1","source_status":200,"source_url":"https://export.arxiv.org/rss/cs.AI","spider":"vendor_press","summary_excerpt":"arXiv:2601.22588v2 Announce Type: replace-cross \nAbstract: Large language models (LLMs) are widely used as reference-free evaluators via prompting, but this \"LLM-as-a-Judge\" paradigm is costly, opaque, and sensitive to prompt design. In this work, we investigate whether smaller models can serve as efficient evaluators by leveraging internal representations instead of surface generation. We uncover a consistent empirical pattern: small LMs, despite with weak generative ability, encode rich evaluative signals in their hidden states. This motivates us to propose the Semantic Capacity Asymmetry Hypothesis: evaluation requires significantly less semantic capacity than generation and can be grounded in intermediate representations, suggesting that evaluation does not necessarily need to rely on large-scale generative models but can instead leverage latent features from smaller ones. Our findings motivate a paradigm shift from LLM-as-a-Judge to Representation-as-a-Judge, a decoding-free evalu","title":"Rethinking LLM-as-a-Judge: Representation-as-a-Judge with Small Language Models via Semantic Capacity Asymmetry","url":"https://arxiv.org/abs/2601.22588","vendor":"arxiv_cs_ai"},"summary":"arXiv:2601.22588v2 Announce Type: replace-cross \nAbstract: Large language models (LLMs) are widely used as reference-free evaluators via prompting, but this \"LLM-as-a-Judge\" paradigm is costly, opaque, and sensitive to prompt design. In this work, we investigate whether smaller models can serve as efficient evaluators by leveraging internal representations instead of surface generation. We uncover a consistent empirical pattern: small LMs, despite with weak generative ability, encode rich evaluative signals in their hidden states. This motivates us to propose the Semantic Capacity Asymmetry Hypothesis: evaluation requires significantly less semantic capacity than generation and can be grounded in intermediate representations, suggesting that evaluation does not necessarily need to rely on large-scale generative models but can instead leverage latent features from smaller ones. Our findings motivate a paradigm shift from LLM-as-a-Judge to Representation-as-a-Judge, a decoding-free evalu","title":"Rethinking LLM-as-a-Judge: Representation-as-a-Judge with Small Language Models via Semantic Capacity Asymmetry","vendor":"arxiv_cs_ai"},"confidence":{"method":"deterministic"},"decision":"POSITIVE","issued_at":"2026-07-10T04:43:53Z","notes":"Spider vendor_press (news) news.vendor_press.v1","object":{"captured_by":"crovia.spider.vendor_press","primary_source_url":"https://arxiv.org/abs/2601.22588"},"predecessors":[],"schema":"crovia.axiom.v1","signature":"ed25519:eea21aec71851d386da28e1d24f0d9f02606b66ecac09cca10195d75a11bda423f63e602b3191a6d843e1d23d23d69205d679da20c8b19b61ba8814e176fe00f","signer":"crovia.substrate","subject":{"observed_at":"2026-07-10T04:43:53Z","source_collector":"spider:vendor_press","target_id":"https://arxiv.org/abs/2601.22588"},"tsa":{"authority":"crovia.substrate.bootstrap","rfc3161_token":"{\"kind\":\"crovia.bootstrap.tsa\",\"source_jsonl\":\"/opt/crovia/spider/data/news/vendor_press_v1.jsonl\",\"source_seal_merkle_root\":\"spider_vendor_press_v1\",\"upgrade_path\":\"Sessione H \\u2014 OpenTimestamps weekly anchor\"}"},"zk_mode":"clear","zk_proof":null},"ledger":{"leaf_hash":"7ca5d80b2882b8a5cc0ed50c891e8c6bcd8c87b2b88387948df1bbe9b9a3c3b9","leaf_index":299552,"ledger_path":"/opt/crovia/substrate/axiom_ledger.jsonl"},"merkle_proof":{"hash_alg":"sha256","leaf_prefix":"0x00","node_prefix":"0x01","odd_leaf_rule":"duplicate_last","path":[{"sibling":"c6e3bbe69502dff06ee000ff93c7e63d6057fa423f6874148ac9968a74e6e1c9","side":"right"},{"sibling":"c6106cb24ba6356779a2a84edb8abae624c427c37acb975b9a88d37810c0c06b","side":"right"},{"sibling":"9da74b50723ae8e0ad02774f45dc8f638fea97749c9c492eb5c8008dc9402eaa","side":"right"},{"sibling":"f30d8faf694d95de590e6b4783242e523cb51190b8e94cae4c546337c0dc8c4b","side":"right"},{"sibling":"14c81aed6356ead3d475bda6b41e856c7bab59b92f6396d2fb9cc2a6e8969fb3","side":"right"},{"sibling":"69bc55ca52b46f95381f84888ef673a19fb31347f7ce79a9f0cd60764972014e","side":"left"},{"sibling":"7e9fd2eac39b5e168659c352c1d3113d1e9ed668787b28903e9c9bddf865d98d","side":"right"},{"sibling":"81caca35f4fef3b5f8e889f7c59fae5c0d4d1710c9d35739fc37260996545e89","side":"right"},{"sibling":"2ca25a5e8823f400322cb0217ea4ad63fb3f05e17cb64a507a02c6bca2008beb","side":"right"},{"sibling":"8200c85c25a948eae071916caaa22b35b677accd5950f58babc0d7656c5f33dc","side":"left"},{"sibling":"f80e8d47e0860527b906fc2dba9a52609f7f623f2772479ae922cc019bab36d9","side":"right"},{"sibling":"cae83500ab2c25555aa6b5eaf9232696d15a868d91b34f7531dd955daadf70f7","side":"right"},{"sibling":"64dab64d51bdcb909e2a5e37efb8909d6704ecf824be484b5d2b60ee6e518890","side":"left"},{"sibling":"576f134a23c19a758ae5efd53016092a74b9900e878cf6eb4f3dab6be682b395","side":"right"},{"sibling":"3c65f53d7c3e4feba7c745e8df1327760ffa768eec84336db14d515a31731532","side":"right"},{"sibling":"d8b9143917b539c543cf4448cec00131f8b807bd8004979c54ebe09798748c66","side":"left"},{"sibling":"8728642cdb98496d916cc653f0d919d7eb2e89d0c529927c9e89091074ad584c","side":"right"},{"sibling":"8025674cb002a22ae243ca0c295c18c1d0ee119189ea88e08ac14a3a1468b8e3","side":"right"},{"sibling":"1cecb7f447febd025aac272837c80de218aecc6485d2395a509b2a1f1b9c746e","side":"left"}]},"schema":"crovia.axiom_proof.v1","seal":{"first_collector_run_id":"","first_receipt_hash":"","jsonl_path":"/opt/crovia/substrate/axiom_ledger.jsonl","key_id":"430895f101d38164","last_collector_run_id":"","last_receipt_hash":"","leaf_count":299721,"merkle_root":"f7115d63193d3285ca28cb9f741ecea2513f9b3e492e785f97076f3cf8f9bb98","public_key_hex":"cf742e26f75669dc673cb5c0786a1ae23ae8ca19c347317192ce40c28a7ff25c","run_id":"hourly_json_retrofit_20260710T053701Z","schema":"crovia.seal.v1","seal_family_version":"crovia-seal-family/1","seal_kind":"substrate_batch","sealed_at":"2026-07-10T05:38:19Z","sig_algorithm":"ed25519","signature":"4eeedeb744885bff6523d66b1cb86bde62b36917696367addfd980d90b31011daece2e01b20608e4c0870ec31dd0bcb57d297447f01f153bf0716ad527142b00","signer_version":"1.1.0"},"trust_root":{"key_id":"430895f101d38164","public_key_hex":"cf742e26f75669dc673cb5c0786a1ae23ae8ca19c347317192ce40c28a7ff25c","signature_algorithm":"ed25519","url":"/registry/canon/TRUST_ROOT.md"},"verifier":{"spec":"/registry/canon/AXIOM_RECEIPT_v1.md","url":"/v/axm_2a61018a51df6a2ee5453ae39a1ad78acf8e622d7875d83f88743d9c9b043b5c"}}