{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:4IK2MKEFW3GQQCBLXRZCYOWYTW","short_pith_number":"pith:4IK2MKEF","canonical_record":{"source":{"id":"2605.23057","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-21T21:46:57Z","cross_cats_sorted":["cs.CL","cs.PF"],"title_canon_sha256":"74bbffdd569781c3ada10f656456ed0c8bb19781c8ddb1518c0cd9b446966bc1","abstract_canon_sha256":"a256308243adf7fbeac5de3cbc57be2b6a0c563251cb82d90c737cda84cd3e8e"},"schema_version":"1.0"},"canonical_sha256":"e215a62885b6cd08082bbc722c3ad89da759214b983b9422ca70d9447d90ea38","source":{"kind":"arxiv","id":"2605.23057","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.23057","created_at":"2026-05-25T02:01:36Z"},{"alias_kind":"arxiv_version","alias_value":"2605.23057v1","created_at":"2026-05-25T02:01:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.23057","created_at":"2026-05-25T02:01:36Z"},{"alias_kind":"pith_short_12","alias_value":"4IK2MKEFW3GQ","created_at":"2026-05-25T02:01:36Z"},{"alias_kind":"pith_short_16","alias_value":"4IK2MKEFW3GQQCBL","created_at":"2026-05-25T02:01:36Z"},{"alias_kind":"pith_short_8","alias_value":"4IK2MKEF","created_at":"2026-05-25T02:01:36Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:4IK2MKEFW3GQQCBLXRZCYOWYTW","target":"record","payload":{"canonical_record":{"source":{"id":"2605.23057","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-21T21:46:57Z","cross_cats_sorted":["cs.CL","cs.PF"],"title_canon_sha256":"74bbffdd569781c3ada10f656456ed0c8bb19781c8ddb1518c0cd9b446966bc1","abstract_canon_sha256":"a256308243adf7fbeac5de3cbc57be2b6a0c563251cb82d90c737cda84cd3e8e"},"schema_version":"1.0"},"canonical_sha256":"e215a62885b6cd08082bbc722c3ad89da759214b983b9422ca70d9447d90ea38","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-25T02:01:36.542390Z","signature_b64":"uh5tHYrqv0ljINNRglH7kuVh9ZX0K6w79+5a3RDRpowEvTysglQZt0gK9Q9zh5xve+pLfVdECce6n67C4yV6Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e215a62885b6cd08082bbc722c3ad89da759214b983b9422ca70d9447d90ea38","last_reissued_at":"2026-05-25T02:01:36.541640Z","signature_status":"signed_v1","first_computed_at":"2026-05-25T02:01:36.541640Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.23057","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-25T02:01:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6IoK6PGIp6OgfVkUq6P7K7ekRiqbIUQJMldCMYO7I7RqD9LS9vazdnUrYT7biCgdNcgqg+ss4KiRE5ETrMVPBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T11:13:46.101910Z"},"content_sha256":"a865411efec65dc9182ba6c887ccc72eef34c9d5a9a26149cb62f36eb27268ce","schema_version":"1.0","event_id":"sha256:a865411efec65dc9182ba6c887ccc72eef34c9d5a9a26149cb62f36eb27268ce"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:4IK2MKEFW3GQQCBLXRZCYOWYTW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ModeSwitch-LLM: A Lightweight Phase-Aware Controller for Cross-Mode LLM Inference on a Single GPU","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.PF"],"primary_cat":"cs.LG","authors_text":"Ali Alshehhi, Aman Sunesh, Hivansh Dhakne","submitted_at":"2026-05-21T21:46:57Z","abstract_excerpt":"ModeSwitch-LLM is a lightweight request-boundary controller for improving single-GPU large language model inference efficiency by routing each request to an appropriate fixed inference mode. Instead of relying on one static serving configuration, the system selects among FP16, quantized modes, speculative decoding, and hybrid modes such as GPTQ plus prefix caching and INT8 plus continuous batching using cheap workload-level features. We evaluate ModeSwitch-LLM on Meta-Llama-3.1-8B-Instruct served on a single NVIDIA A100 GPU. On deployment-style synthetic workloads, the online controller achiev"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.23057","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/2605.23057/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-25T02:01:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SBNIG2+Q/vX4W7E0LC/1yq7ruvUzqKChvVXhVM8sJOeXP9fWhS2Hz18LfJCN4epYqAXUBN/qxltHUMqfUEtjCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T11:13:46.102682Z"},"content_sha256":"bfa1c3b20932f72918c5e6d1d5180d30a1df779c7ba94e54df5a6b08ba0193b9","schema_version":"1.0","event_id":"sha256:bfa1c3b20932f72918c5e6d1d5180d30a1df779c7ba94e54df5a6b08ba0193b9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4IK2MKEFW3GQQCBLXRZCYOWYTW/bundle.json","state_url":"https://pith.science/pith/4IK2MKEFW3GQQCBLXRZCYOWYTW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4IK2MKEFW3GQQCBLXRZCYOWYTW/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-10T11:13:46Z","links":{"resolver":"https://pith.science/pith/4IK2MKEFW3GQQCBLXRZCYOWYTW","bundle":"https://pith.science/pith/4IK2MKEFW3GQQCBLXRZCYOWYTW/bundle.json","state":"https://pith.science/pith/4IK2MKEFW3GQQCBLXRZCYOWYTW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4IK2MKEFW3GQQCBLXRZCYOWYTW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:4IK2MKEFW3GQQCBLXRZCYOWYTW","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":"a256308243adf7fbeac5de3cbc57be2b6a0c563251cb82d90c737cda84cd3e8e","cross_cats_sorted":["cs.CL","cs.PF"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-21T21:46:57Z","title_canon_sha256":"74bbffdd569781c3ada10f656456ed0c8bb19781c8ddb1518c0cd9b446966bc1"},"schema_version":"1.0","source":{"id":"2605.23057","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.23057","created_at":"2026-05-25T02:01:36Z"},{"alias_kind":"arxiv_version","alias_value":"2605.23057v1","created_at":"2026-05-25T02:01:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.23057","created_at":"2026-05-25T02:01:36Z"},{"alias_kind":"pith_short_12","alias_value":"4IK2MKEFW3GQ","created_at":"2026-05-25T02:01:36Z"},{"alias_kind":"pith_short_16","alias_value":"4IK2MKEFW3GQQCBL","created_at":"2026-05-25T02:01:36Z"},{"alias_kind":"pith_short_8","alias_value":"4IK2MKEF","created_at":"2026-05-25T02:01:36Z"}],"graph_snapshots":[{"event_id":"sha256:bfa1c3b20932f72918c5e6d1d5180d30a1df779c7ba94e54df5a6b08ba0193b9","target":"graph","created_at":"2026-05-25T02:01:36Z","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/2605.23057/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"ModeSwitch-LLM is a lightweight request-boundary controller for improving single-GPU large language model inference efficiency by routing each request to an appropriate fixed inference mode. Instead of relying on one static serving configuration, the system selects among FP16, quantized modes, speculative decoding, and hybrid modes such as GPTQ plus prefix caching and INT8 plus continuous batching using cheap workload-level features. We evaluate ModeSwitch-LLM on Meta-Llama-3.1-8B-Instruct served on a single NVIDIA A100 GPU. On deployment-style synthetic workloads, the online controller achiev","authors_text":"Ali Alshehhi, Aman Sunesh, Hivansh Dhakne","cross_cats":["cs.CL","cs.PF"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-21T21:46:57Z","title":"ModeSwitch-LLM: A Lightweight Phase-Aware Controller for Cross-Mode LLM Inference on a Single GPU"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.23057","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:a865411efec65dc9182ba6c887ccc72eef34c9d5a9a26149cb62f36eb27268ce","target":"record","created_at":"2026-05-25T02:01:36Z","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":"a256308243adf7fbeac5de3cbc57be2b6a0c563251cb82d90c737cda84cd3e8e","cross_cats_sorted":["cs.CL","cs.PF"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-21T21:46:57Z","title_canon_sha256":"74bbffdd569781c3ada10f656456ed0c8bb19781c8ddb1518c0cd9b446966bc1"},"schema_version":"1.0","source":{"id":"2605.23057","kind":"arxiv","version":1}},"canonical_sha256":"e215a62885b6cd08082bbc722c3ad89da759214b983b9422ca70d9447d90ea38","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e215a62885b6cd08082bbc722c3ad89da759214b983b9422ca70d9447d90ea38","first_computed_at":"2026-05-25T02:01:36.541640Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-25T02:01:36.541640Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"uh5tHYrqv0ljINNRglH7kuVh9ZX0K6w79+5a3RDRpowEvTysglQZt0gK9Q9zh5xve+pLfVdECce6n67C4yV6Cw==","signature_status":"signed_v1","signed_at":"2026-05-25T02:01:36.542390Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.23057","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a865411efec65dc9182ba6c887ccc72eef34c9d5a9a26149cb62f36eb27268ce","sha256:bfa1c3b20932f72918c5e6d1d5180d30a1df779c7ba94e54df5a6b08ba0193b9"],"state_sha256":"3ad4d4153b1560e01a257f157c9dac77b2789d5120aa86788c08406af15863cc"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xUjZ9tsGySHhDlgEFUjhFn2/ietidvavUWLdcvCTyf4s42Gs4vfeZ3I/OUcOBw1+tbzQeR6qbdS5vbdImGF6AA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-10T11:13:46.106712Z","bundle_sha256":"3abb58e011290f3347ddfd94a79d7bb5e5b4fc71a79d1bba452faa7952d35237"}}