{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:H4NDJZ4N7A2T7ZRFFDA6H3YAQW","short_pith_number":"pith:H4NDJZ4N","schema_version":"1.0","canonical_sha256":"3f1a34e78df8353fe62528c1e3ef008582fc87aa81a76b9234a849a9e0a02cfe","source":{"kind":"arxiv","id":"2608.07427","version":1},"attestation_state":"computed","paper":{"title":"A Picture is Worth a Thousand Tokens: How Vision Language Models Cut AI Energy Costs While Improving Accuracy","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.PF"],"primary_cat":"cs.AI","authors_text":"Bhavika Jalli, Jayanta Choudhury, Nikhil Korati Prasanna","submitted_at":"2026-08-07T17:14:45Z","abstract_excerpt":"LLM inference accounts for over 90% of AI operational energy, scaling directly with input token count---a critical inefficiency for telecom network analytics and numerical time-series data analysis (NTSDA), where raw multivariate KPI windows from 4G/5G cell sites expand into thousands of floating-point tokens. Vision-Language Models (VLMs) eliminate this mismatch by encoding time-series as 2D plots, achieving 3.6-10.4x input token reduction across Llama-3.2-90B, Qwen2.5-VL-72B, and Pixtral-12B architectures. This translates to 1.8-2.5x measured inference energy reduction, saving approximately "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2608.07427","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-07T17:14:45Z","cross_cats_sorted":["cs.PF"],"title_canon_sha256":"1ac97aba2ece7daa227a18496dee90826c0f96b3eb324482f7f7e1d8d37dbed1","abstract_canon_sha256":"05260dbe5b63e05d11c6e8de997f06fbc431b52386943febd90a0378912939fd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-10T01:15:22.113265Z","signature_b64":"6Wvdd8P2z8a1sismWEdsyflK/zf486h57auMHjSZ/lgcIPmlfTVAQzh9utqXSJR4wZl4Ixs3Q/+zahVc5pOuCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3f1a34e78df8353fe62528c1e3ef008582fc87aa81a76b9234a849a9e0a02cfe","last_reissued_at":"2026-08-10T01:15:22.110911Z","signature_status":"signed_v1","first_computed_at":"2026-08-10T01:15:22.110911Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Picture is Worth a Thousand Tokens: How Vision Language Models Cut AI Energy Costs While Improving Accuracy","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.PF"],"primary_cat":"cs.AI","authors_text":"Bhavika Jalli, Jayanta Choudhury, Nikhil Korati Prasanna","submitted_at":"2026-08-07T17:14:45Z","abstract_excerpt":"LLM inference accounts for over 90% of AI operational energy, scaling directly with input token count---a critical inefficiency for telecom network analytics and numerical time-series data analysis (NTSDA), where raw multivariate KPI windows from 4G/5G cell sites expand into thousands of floating-point tokens. Vision-Language Models (VLMs) eliminate this mismatch by encoding time-series as 2D plots, achieving 3.6-10.4x input token reduction across Llama-3.2-90B, Qwen2.5-VL-72B, and Pixtral-12B architectures. This translates to 1.8-2.5x measured inference energy reduction, saving approximately "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.07427","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2608.07427/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2608.07427","created_at":"2026-08-10T01:15:22.111880+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.07427v1","created_at":"2026-08-10T01:15:22.111880+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.07427","created_at":"2026-08-10T01:15:22.111880+00:00"},{"alias_kind":"pith_short_12","alias_value":"H4NDJZ4N7A2T","created_at":"2026-08-10T01:15:22.111880+00:00"},{"alias_kind":"pith_short_16","alias_value":"H4NDJZ4N7A2T7ZRF","created_at":"2026-08-10T01:15:22.111880+00:00"},{"alias_kind":"pith_short_8","alias_value":"H4NDJZ4N","created_at":"2026-08-10T01:15:22.111880+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/H4NDJZ4N7A2T7ZRFFDA6H3YAQW","json":"https://pith.science/pith/H4NDJZ4N7A2T7ZRFFDA6H3YAQW.json","graph_json":"https://pith.science/api/pith-number/H4NDJZ4N7A2T7ZRFFDA6H3YAQW/graph.json","events_json":"https://pith.science/api/pith-number/H4NDJZ4N7A2T7ZRFFDA6H3YAQW/events.json","paper":"https://pith.science/paper/H4NDJZ4N"},"agent_actions":{"view_html":"https://pith.science/pith/H4NDJZ4N7A2T7ZRFFDA6H3YAQW","download_json":"https://pith.science/pith/H4NDJZ4N7A2T7ZRFFDA6H3YAQW.json","view_paper":"https://pith.science/paper/H4NDJZ4N","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.07427&json=true","fetch_graph":"https://pith.science/api/pith-number/H4NDJZ4N7A2T7ZRFFDA6H3YAQW/graph.json","fetch_events":"https://pith.science/api/pith-number/H4NDJZ4N7A2T7ZRFFDA6H3YAQW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/H4NDJZ4N7A2T7ZRFFDA6H3YAQW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/H4NDJZ4N7A2T7ZRFFDA6H3YAQW/action/storage_attestation","attest_author":"https://pith.science/pith/H4NDJZ4N7A2T7ZRFFDA6H3YAQW/action/author_attestation","sign_citation":"https://pith.science/pith/H4NDJZ4N7A2T7ZRFFDA6H3YAQW/action/citation_signature","submit_replication":"https://pith.science/pith/H4NDJZ4N7A2T7ZRFFDA6H3YAQW/action/replication_record"}},"created_at":"2026-08-10T01:15:22.111880+00:00","updated_at":"2026-08-10T01:15:22.111880+00:00"}