{"id":"https://openalex.org/W7162567538","doi":"https://doi.org/10.48550/arxiv.2605.27154","title":"Touch-R1: Reinforcing Touch Reasoning in MLLMs","display_name":"Touch-R1: Reinforcing Touch Reasoning in MLLMs","publication_year":2026,"publication_date":"2026-05-26","ids":{"openalex":"https://openalex.org/W7162567538","doi":"https://doi.org/10.48550/arxiv.2605.27154"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.27154","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.27154","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.2605.27154","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5137180718","display_name":"Yingxin Lai","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lai, Yingxin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137111782","display_name":"Yafei Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Yafei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137158554","display_name":"Fucai Zhu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhu, Fucai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030051087","display_name":"Siyu Zhu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhu, Siyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5137147419","display_name":"Weihao Yuan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuan, Weihao","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/T10338","display_name":"Advanced Sensor and Energy Harvesting Materials","score":0.6929000020027161,"subfield":{"id":"https://openalex.org/subfields/2204","display_name":"Biomedical Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10338","display_name":"Advanced Sensor and Energy Harvesting Materials","score":0.6929000020027161,"subfield":{"id":"https://openalex.org/subfields/2204","display_name":"Biomedical 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/T10914","display_name":"Tactile and Sensory Interactions","score":0.21639999747276306,"subfield":{"id":"https://openalex.org/subfields/2805","display_name":"Cognitive Neuroscience"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T12611","display_name":"Neural Networks and Reservoir Computing","score":0.01360000018030405,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/counterfactual-thinking","display_name":"Counterfactual thinking","score":0.5669999718666077},{"id":"https://openalex.org/keywords/perception","display_name":"Perception","score":0.5324000120162964},{"id":"https://openalex.org/keywords/correctness","display_name":"Correctness","score":0.520799994468689},{"id":"https://openalex.org/keywords/embodied-cognition","display_name":"Embodied cognition","score":0.3546999990940094},{"id":"https://openalex.org/keywords/tactile-stimuli","display_name":"Tactile stimuli","score":0.3328000009059906},{"id":"https://openalex.org/keywords/tactile-sensor","display_name":"Tactile sensor","score":0.326200008392334},{"id":"https://openalex.org/keywords/visual-perception","display_name":"Visual perception","score":0.32269999384880066}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5769000053405762},{"id":"https://openalex.org/C108650721","wikidata":"https://www.wikidata.org/wiki/Q1783253","display_name":"Counterfactual thinking","level":2,"score":0.5669999718666077},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.559499979019165},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.5324000120162964},{"id":"https://openalex.org/C55439883","wikidata":"https://www.wikidata.org/wiki/Q360812","display_name":"Correctness","level":2,"score":0.520799994468689},{"id":"https://openalex.org/C100609095","wikidata":"https://www.wikidata.org/wiki/Q1335050","display_name":"Embodied cognition","level":2,"score":0.3546999990940094},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.3375999927520752},{"id":"https://openalex.org/C2987654038","wikidata":"https://www.wikidata.org/wiki/Q859031","display_name":"Tactile stimuli","level":3,"score":0.3328000009059906},{"id":"https://openalex.org/C46722567","wikidata":"https://www.wikidata.org/wiki/Q7674139","display_name":"Tactile sensor","level":3,"score":0.326200008392334},{"id":"https://openalex.org/C178253425","wikidata":"https://www.wikidata.org/wiki/Q162668","display_name":"Visual perception","level":3,"score":0.32269999384880066},{"id":"https://openalex.org/C3017819093","wikidata":"https://www.wikidata.org/wiki/Q328835","display_name":"Tactile perception","level":3,"score":0.32249999046325684},{"id":"https://openalex.org/C97364631","wikidata":"https://www.wikidata.org/wiki/Q484284","display_name":"Deductive reasoning","level":2,"score":0.32100000977516174},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.3149999976158142},{"id":"https://openalex.org/C180747234","wikidata":"https://www.wikidata.org/wiki/Q23373","display_name":"Cognitive psychology","level":1,"score":0.31459999084472656},{"id":"https://openalex.org/C193221554","wikidata":"https://www.wikidata.org/wiki/Q5153664","display_name":"Commonsense reasoning","level":2,"score":0.3070000112056732},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.28700000047683716},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.2822999954223633},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.26980000734329224},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.2635999917984009},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.25369998812675476}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.27154","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.27154","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.2605.27154","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.27154","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":[{"score":0.5170838832855225,"id":"https://metadata.un.org/sdg/16","display_name":"Peace, Justice and strong institutions"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"While":[0],"rule-based":[1],"reinforcement":[2],"learning":[3],"has":[4],"recently":[5],"catalyzed":[6],"explicit":[7],"reasoning":[8,13,43,108,178,191],"in":[9,34,63,196],"multimodal":[10,75],"models,":[11],"tactile":[12,65,81,94,107,133,145,156],"remains":[14],"largely":[15],"underexplored.":[16],"Existing":[17],"tactile-language":[18],"models":[19],"primarily":[20],"rely":[21],"on":[22,111,174],"supervised":[23],"or":[24,37,161],"contrastive":[25],"objectives,":[26],"which":[27],"limits":[28],"their":[29],"capacity":[30],"to":[31,151],"ground":[32],"predictions":[33],"physical":[35,52,126,197],"evidence":[36],"rectify":[38],"misleading":[39],"visual":[40],"priors.":[41],"Tactile":[42],"introduces":[44],"two":[45],"modality-specific":[46],"challenges:":[47],"the":[48,58,137,155],"ordinal":[49],"nature":[50],"of":[51,183],"attributes":[53],"(e.g.,":[54],"hardness,":[55],"roughness)":[56],"and":[57,87,96,130,170,186],"cross-sensor":[59,125],"distribution":[60],"shifts":[61],"inherent":[62],"optical":[64],"hardware.":[66],"In":[67],"this":[68],"work,":[69],"we":[70,103],"introduce":[71],"TouchReason-1M,":[72],"a":[73,89,106,117],"large-scale":[74],"dataset":[76],"comprising":[77],"over":[78],"1M":[79],"synchronized":[80],"pairs":[82],"across":[83],"four":[84],"distinct":[85],"sensors,":[86],"TouchReason-Bench,":[88,164],"rigorous":[90],"framework":[91],"for":[92],"evaluating":[93],"perception":[95],"visual-tactile":[97],"conflict":[98],"resolution.":[99],"Building":[100],"upon":[101],"these,":[102],"propose":[104],"Touch-R1,":[105],"MLLM":[109],"based":[110],"Qwen2.5-VL-7B.":[112],"Touch-R1":[113],"is":[114,158],"trained":[115],"via":[116],"tactile-grounded":[118],"GRPO":[119],"objective":[120],"that":[121,189],"combines":[122],"ordinal-aware":[123],"accuracy,":[124],"consistency,":[127],"structured-format":[128],"control,":[129],"an":[131],"input-side":[132],"grounding":[134],"objective.":[135],"Specifically,":[136],"tactile-use":[138],"reward":[139],"assigns":[140],"credit":[141],"only":[142],"when":[143],"authentic":[144],"inputs":[146],"yield":[147],"superior":[148],"correctness":[149],"relative":[150],"counterfactual":[152],"controls":[153],"where":[154],"stream":[157],"removed,":[159],"shuffled,":[160],"noise-masked.":[162],"On":[163],"Touch-R1-7B":[165],"outperforms":[166],"Octopi-13B":[167],"by":[168,172],"18.4\\%":[169],"GPT-4o":[171],"24.7\\%":[173],"average.":[175],"Its":[176],"structured":[177],"traces":[179],"reveal":[180],"emergent":[181],"behaviors":[182],"probing,":[184],"comparison,":[185],"revision,":[187],"demonstrating":[188],"R1-style":[190],"can":[192],"be":[193],"effectively":[194],"grounded":[195],"contact.":[198]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-28T00:00:00"}