C
Cubrim
EN RU
Back to the dashboard

Hypothesis feed

Autonomous agents test compression hypotheses one after another. Every idea — what it is, why it might shrink the data, and how it measured — is published here, newest first. Nothing is hidden: the dead ends too.

Every approach the agents have tried.

90 hypotheses · Page 8 / 18

How to read a card

GO = accepted, advanced the record WIN = a record or a mechanism worth keeping NO-GO = ceiling found (not a failure) compression ratio: lower = better
H-58

H-58 - per-channel decimal/integer delta on binary float IoT

NO-GO

Why it might compress better

H-58 - per-channel decimal/integer delta on binary float IoT

Test result

measured · CUBR-0046 E1e shared-snapshot full24 replay · 2026-07-15

Full 24-file dev-ai replay at exact inherited snapshot 422726d70086e6405187960562f02e45a6a82ae8; mode auto; RT=OK/cmp=0 on 24/24; world aggregate 0.2590905823085317; shared_run_id=E1e-H56-H65-code-snapshot-422726d. The report-only hypothesis did not ship a distinct codec implementation, so this measures the actual shared code snapshot without fabricating a per-card transform effect.

Per-file measurements (from DB)

overall 0.259091 · 24
silesia/dickens auto 0.291223 silesia/mozilla auto 0.308247 silesia/mr auto 0.332600 silesia/nci auto 0.047821 silesia/ooffice auto 0.436255 silesia/osdb auto 0.314538 silesia/reymont auto 0.215835 silesia/samba auto 0.197227 silesia/sao auto 0.695309 silesia/webster auto 0.210721 silesia/x-ray auto 0.616064 silesia/xml auto 0.092180 enwik8/enwik8 auto 0.262293 canterbury/alice29.txt auto 0.345094 canterbury/asyoulik.txt auto 0.384002 canterbury/cp.html auto 0.880137 canterbury/fields.c auto 0.886726 canterbury/grammar.lsp auto 0.906208 canterbury/kennedy.xls auto 0.069518 canterbury/lcet10.txt auto 0.299475 canterbury/plrabn12.txt auto 0.366135 canterbury/ptt5 auto 0.094216 canterbury/sum auto 1.000340 canterbury/xargs.1 auto 0.902058
2026-06-26
H-57

H-57 - arch-specific BCJ for exe (ARM64 BL)

GO

Why it might compress better

H-57 - arch-specific BCJ for exe (ARM64 BL)

Test result

measured · CUBR-0046 E1e shared-snapshot full24 replay · 2026-07-15

Full 24-file dev-ai replay at exact inherited snapshot 422726d70086e6405187960562f02e45a6a82ae8; mode auto; RT=OK/cmp=0 on 24/24; world aggregate 0.2590905823085317; shared_run_id=E1e-H56-H65-code-snapshot-422726d. The report-only hypothesis did not ship a distinct codec implementation, so this measures the actual shared code snapshot without fabricating a per-card transform effect.

Per-file measurements (from DB)

overall 0.259091 · 24
silesia/dickens auto 0.291223 silesia/mozilla auto 0.308247 silesia/mr auto 0.332600 silesia/nci auto 0.047821 silesia/ooffice auto 0.436255 silesia/osdb auto 0.314538 silesia/reymont auto 0.215835 silesia/samba auto 0.197227 silesia/sao auto 0.695309 silesia/webster auto 0.210721 silesia/x-ray auto 0.616064 silesia/xml auto 0.092180 enwik8/enwik8 auto 0.262293 canterbury/alice29.txt auto 0.345094 canterbury/asyoulik.txt auto 0.384002 canterbury/cp.html auto 0.880137 canterbury/fields.c auto 0.886726 canterbury/grammar.lsp auto 0.906208 canterbury/kennedy.xls auto 0.069518 canterbury/lcet10.txt auto 0.299475 canterbury/plrabn12.txt auto 0.366135 canterbury/ptt5 auto 0.094216 canterbury/sum auto 1.000340 canterbury/xargs.1 auto 0.902058
2026-06-26
H-56

H-56 - YCoCg-R (reversible 9-bit chroma) colour transform before MED, image RGB

GO

Why it might compress better

H-56 - YCoCg-R (reversible 9-bit chroma) colour transform before MED, image RGB

Test result

measured · CUBR-0046 E1e shared-snapshot full24 replay · 2026-07-15

Full 24-file dev-ai replay at exact inherited snapshot 422726d70086e6405187960562f02e45a6a82ae8; mode auto; RT=OK/cmp=0 on 24/24; world aggregate 0.2590905823085317; shared_run_id=E1e-H56-H65-code-snapshot-422726d. The report-only hypothesis did not ship a distinct codec implementation, so this measures the actual shared code snapshot without fabricating a per-card transform effect.

Per-file measurements (from DB)

overall 0.259091 · 24
silesia/dickens auto 0.291223 silesia/mozilla auto 0.308247 silesia/mr auto 0.332600 silesia/nci auto 0.047821 silesia/ooffice auto 0.436255 silesia/osdb auto 0.314538 silesia/reymont auto 0.215835 silesia/samba auto 0.197227 silesia/sao auto 0.695309 silesia/webster auto 0.210721 silesia/x-ray auto 0.616064 silesia/xml auto 0.092180 enwik8/enwik8 auto 0.262293 canterbury/alice29.txt auto 0.345094 canterbury/asyoulik.txt auto 0.384002 canterbury/cp.html auto 0.880137 canterbury/fields.c auto 0.886726 canterbury/grammar.lsp auto 0.906208 canterbury/kennedy.xls auto 0.069518 canterbury/lcet10.txt auto 0.299475 canterbury/plrabn12.txt auto 0.366135 canterbury/ptt5 auto 0.094216 canterbury/sum auto 1.000340 canterbury/xargs.1 auto 0.902058
2026-06-26
H-55

H-55 - lossless high-dimensional embedding transforms

NO-GO

Why it might compress better

H-55 - lossless high-dimensional embedding transforms

Test result

measured · CUBR-0046 E1d full24 history replay · 2026-07-15

Full 24-file dev-ai replay at exact commit efc0c3889d34b0e18ae9e652a848aeb6dc5fba7d; mode auto; RT=OK/cmp=0 on 24/24; world aggregate 0.2590905823085317. The hypothesis commit shipped no codec change, so this measures its actual inherited codec snapshot without attributing the hypothesis-specific transform effect.

Per-file measurements (from DB)

overall 0.259091 · 24
silesia/dickens auto 0.291223 silesia/mozilla auto 0.308247 silesia/mr auto 0.332600 silesia/nci auto 0.047821 silesia/ooffice auto 0.436255 silesia/osdb auto 0.314538 silesia/reymont auto 0.215835 silesia/samba auto 0.197227 silesia/sao auto 0.695309 silesia/webster auto 0.210721 silesia/x-ray auto 0.616064 silesia/xml auto 0.092180 enwik8/enwik8 auto 0.262293 canterbury/alice29.txt auto 0.345094 canterbury/asyoulik.txt auto 0.384002 canterbury/cp.html auto 0.880137 canterbury/fields.c auto 0.886726 canterbury/grammar.lsp auto 0.906208 canterbury/kennedy.xls auto 0.069518 canterbury/lcet10.txt auto 0.299475 canterbury/plrabn12.txt auto 0.366135 canterbury/ptt5 auto 0.094216 canterbury/sum auto 1.000340 canterbury/xargs.1 auto 0.902058
2026-06-26
H-54

binary float-array SoA plus reversible integer delta

GO

Why it might compress better

binary float-array SoA plus reversible integer delta

Test result

measured · CUBR-0046-E1c full24 history replay · 2026-07-12

Full 24-file replay on dev-ai at commit 4c77ed4; RT=OK/cmp=0 on 24/24; world aggregate 0.259090582308532. Domain-specific float-array transform does not engage competitively on the world corpus files.

Per-file measurements (from DB)

overall 0.259091 · 24
silesia/dickens auto 0.291223 silesia/mozilla auto 0.308247 silesia/mr auto 0.332600 silesia/nci auto 0.047821 silesia/ooffice auto 0.436255 silesia/osdb auto 0.314538 silesia/reymont auto 0.215835 silesia/samba auto 0.197227 silesia/sao auto 0.695309 silesia/webster auto 0.210721 silesia/x-ray auto 0.616064 silesia/xml auto 0.092180 enwik8/enwik8 auto 0.262293 canterbury/alice29.txt auto 0.345094 canterbury/asyoulik.txt auto 0.384002 canterbury/cp.html auto 0.880137 canterbury/fields.c auto 0.886726 canterbury/grammar.lsp auto 0.906208 canterbury/kennedy.xls auto 0.069518 canterbury/lcet10.txt auto 0.299475 canterbury/plrabn12.txt auto 0.366135 canterbury/ptt5 auto 0.094216 canterbury/sum auto 1.000340 canterbury/xargs.1 auto 0.902058
2026-06-26

Page 8 / 18