{"id":"https://openalex.org/W7163181954","doi":"https://doi.org/10.48550/arxiv.2606.00611","title":"TRACE: Trajectory Risk-Aware Compression for Long-Horizon Agent Safety","display_name":"TRACE: Trajectory Risk-Aware Compression for Long-Horizon Agent Safety","publication_year":2026,"publication_date":"2026-05-30","ids":{"openalex":"https://openalex.org/W7163181954","doi":"https://doi.org/10.48550/arxiv.2606.00611"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.00611","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.00611","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.00611","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5124186144","display_name":"Zhepei Hong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hong, Zhepei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137633170","display_name":"Lin Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Lin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137700831","display_name":"Liting Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Liting","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5006833163","display_name":"Haokai Ma","orcid":"https://orcid.org/0000-0002-4621-5213"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ma, Haokai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137700727","display_name":"Junfeng Fang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fang, Junfeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137681668","display_name":"Fei Shen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shen, Fei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137654915","display_name":"Dan Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Dan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5137644321","display_name":"Xiang Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Xiang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.37709999084472656,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.37709999084472656,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11099","display_name":"Autonomous Vehicle Technology and Safety","score":0.13899999856948853,"subfield":{"id":"https://openalex.org/subfields/2203","display_name":"Automotive Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11622","display_name":"Maritime Navigation and Safety","score":0.09049999713897705,"subfield":{"id":"https://openalex.org/subfields/2212","display_name":"Ocean Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/trajectory","display_name":"Trajectory","score":0.8051999807357788},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.6110000014305115},{"id":"https://openalex.org/keywords/trace","display_name":"TRACE (psycholinguistics)","score":0.5795000195503235},{"id":"https://openalex.org/keywords/focus","display_name":"Focus (optics)","score":0.5253000259399414},{"id":"https://openalex.org/keywords/compression","display_name":"Compression (physics)","score":0.5152000188827515},{"id":"https://openalex.org/keywords/aggregate","display_name":"Aggregate (composite)","score":0.5049999952316284},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.4246000051498413}],"concepts":[{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.8051999807357788},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7092000246047974},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.6110000014305115},{"id":"https://openalex.org/C75291252","wikidata":"https://www.wikidata.org/wiki/Q1315756","display_name":"TRACE (psycholinguistics)","level":2,"score":0.5795000195503235},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.5253000259399414},{"id":"https://openalex.org/C180016635","wikidata":"https://www.wikidata.org/wiki/Q2712821","display_name":"Compression (physics)","level":2,"score":0.5152000188827515},{"id":"https://openalex.org/C4679612","wikidata":"https://www.wikidata.org/wiki/Q866298","display_name":"Aggregate (composite)","level":2,"score":0.5049999952316284},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.4246000051498413},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.38260000944137573},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.3734999895095825},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.3197000026702881},{"id":"https://openalex.org/C81081738","wikidata":"https://www.wikidata.org/wiki/Q55542","display_name":"Lossless compression","level":3,"score":0.3082999885082245},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.30799999833106995},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.3043000102043152},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.2953000068664551},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2831999957561493},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2653000056743622},{"id":"https://openalex.org/C2779662365","wikidata":"https://www.wikidata.org/wiki/Q5416694","display_name":"Event (particle physics)","level":2,"score":0.26460000872612},{"id":"https://openalex.org/C44154836","wikidata":"https://www.wikidata.org/wiki/Q45045","display_name":"Simulation","level":1,"score":0.25999999046325684},{"id":"https://openalex.org/C20556612","wikidata":"https://www.wikidata.org/wiki/Q4469374","display_name":"Volume (thermodynamics)","level":2,"score":0.2538999915122986}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.00611","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.00611","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.00611","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.00611","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Long-horizon":[0],"LLM":[1],"agents":[2],"produce":[3],"safety":[4,40,90],"evidence":[5,32,44,71,86,102],"across":[6],"long":[7],"trajectories,":[8],"where":[9],"sparse,":[10],"delayed,":[11],"and":[12,29,46,76,99,107,141,156],"compositional":[13],"risk":[14,97],"signals":[15],"often":[16],"escape":[17],"local":[18],"moderation.":[19],"Existing":[20],"turn-level":[21],"or":[22],"short-context":[23],"detectors":[24],"struggle":[25],"to":[26,124],"reliably":[27],"retain":[28],"aggregate":[30,95],"such":[31],"over":[33,119],"extended":[34],"horizons.":[35],"We":[36],"reframe":[37],"long-horizon":[38],"agent":[39],"detection":[41],"as":[42,88,135],"trajectory-level":[43,74],"compression":[45],"propose":[47],"Trajectory":[48],"Risk-Aware":[49],"Compression":[50],"for":[51],"Long-Horizon":[52],"Agent":[53],"Safety":[54],"(TRACE).":[55],"TRACE":[56,109,130],"uses":[57],"a":[58,68,89],"Compressor-Reader":[59],"design:":[60],"the":[61,64,77,80,111,146,150],"Compressor":[62],"encodes":[63],"full":[65],"trajectory":[66,82],"into":[67],"compact":[69],"latent":[70,85],"state":[72,87],"under":[73],"supervision,":[75],"Reader":[78,151],"judges":[79],"raw":[81],"with":[83],"this":[84],"reference.":[91],"This":[92],"design":[93],"helps":[94,149],"dispersed":[96],"cues":[98],"reduce":[100],"premature":[101],"loss.":[103],"Across":[104],"ASSEBench,":[105],"Pre-Ex-Bench,":[106],"R-Judge,":[108],"achieves":[110],"best":[112],"accuracy":[113],"on":[114,153],"all":[115],"evaluated":[116],"backbones,":[117],"improving":[118],"strong":[120],"baselines":[121],"by":[122],"up":[123],"12.6":[125],"percentage":[126],"points.":[127],"On":[128],"LongSafety,":[129],"shows":[131],"smaller":[132],"performance":[133],"degradation":[134],"context":[136],"length":[137],"grows.":[138],"Attention":[139],"visualizations":[140],"case":[142],"studies":[143],"suggest":[144],"that":[145],"compressed":[147],"reference":[148],"focus":[152],"risk-critical":[154],"segments":[155],"recover":[157],"cross-step":[158],"evidence.":[159],"Code":[160],"is":[161],"available":[162],"at":[163],"https://github.com/Peregrine123/TRACE_official.":[164]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-03T00:00:00"}