Dev claim · claim-478 · #134 of 154
“Protecting and restoring forests would reduce 18% of emissions by 2030”
Claim tags (sorted stems; highlighted when a passage shares them)
- emiss
- forest
- protect
- reduc
- restor
- would
The model’s verdict
The Transformer retrained from the notebook (the 2024 weights were never saved), shown three ways. The first is how the notebook evaluated it.
Notebook protocol
Dev claims predicted in batches of 16, in file order, from the 2024 retrieved evidence, as the notebook does.
- Supports
- 9.3%
- Refutes
- 2.2%
- Not enough info
- 85.3%
- Disputed
- 3.2%
One claim at a time
The same model run on a single claim, which is how the Try-it page runs it.
- Supports
- 31.4%
- Refutes
- 7.4%
- Not enough info
- 49.2%
- Disputed
- 12.0%
With gold evidence
Fed the human-annotated evidence instead of retrieved passages. This is the 'val accuracy' the training loop reports.
- Supports
- 19.4%
- Refutes
- 4.7%
- Not enough info
- 67.2%
- Disputed
- 8.7%
Retrieved vs gold evidence
Gold passages were picked by the dataset’s annotators. Retrieved passages are what the TF-IDF rule returned from all 1.19M passages. Switch between the submitted 2024 output and the re-runs.
2024 submission
The passages the team actually retrieved and submitted in 2024 (their saved output file). The file records only passage ids, so scores are recomputed with the submission rule.
- Path
- not recorded
- Found
- 0 of 5 gold
- P
- 0.00
- R
- 0.00
- F
- 0.00
#1evidence-951517 Its natural habitats are subtropical or tropical moist lowland forests, plantations, rural gardens, urban areas, and heavily degraded former forest.
- forest (shared with the claim)
- cos
- 0.497
- overlap
- 1.000
- score
- 1.497
- shared
- 1
#2evidence-117975 Its natural habitats are temperate forests, subtropical or tropical moist lowland forests, and heavily degraded former forest.
- forest (shared with the claim)
- cos
- 0.497
- overlap
- 1.000
- score
- 1.497
- shared
- 1
#3evidence-367506 Its natural habitats are subtropical or tropical moist lowland forests, intermittent freshwater marshes, and heavily degraded former forest.
- forest (shared with the claim)
- cos
- 0.497
- overlap
- 1.000
- score
- 1.497
- shared
- 1
#4evidence-282339 Its natural habitats are subtropical or tropical moist lowland forests and heavily degraded former forest.
- forest (shared with the claim)
- cos
- 0.497
- overlap
- 1.000
- score
- 1.497
- shared
- 1
#5evidence-75103 Its natural habitats are temperate forests and subtropical or tropical moist lowland forests.
- forest (shared with the claim)
- cos
- 0.497
- overlap
- 1.000
- score
- 1.497
- shared
- 1
#6evidence-33952 Its natural habitats are subtropical or tropical moist lowland forests, rivers, swamps, and heavily degraded former forest.
- forest (shared with the claim)
- cos
- 0.497
- overlap
- 1.000
- score
- 1.497
- shared
- 1
Gold evidence (5)
evidence-890810 10% per annum until zero emissions are reached around 2030.
- reach
- per
- around
- emiss (shared with the claim)
- zero
evidence-328157 Targets for the year 2030: Reduce GHG emission by 40% from the level of 1990.
- level
- emiss (shared with the claim)
- ghg
- target
- reduc (shared with the claim)
- year
evidence-347403 The UN estimate deforestation and forest degradation to make up 17% of global carbon emissions, which makes it the second most polluting sector, following the energy industry.
- make
- degrad
- un
- emiss (shared with the claim)
- deforest
- sector
- pollut
- estim
evidence-594952 Japan needs to reduce 26% of green house gas emission from 2013 by 2030 to accomplish Paris Agreement and is trying to reduce 2% of them by forestry.
- emiss (shared with the claim)
- agreement
- accomplish
- pari
- reduc (shared with the claim)
- ga
- tri
- forestri
evidence-609211 The report mentioned that this would require global net human-caused emissions of carbon dioxide (CO2) to fall by about 45% from 2010 levels by 2030, reaching "net zero" around 2050, through “rapid and far-reaching” transitions in land, energy, industry, buildings, transport, and cities.
- reach
- rapid
- mention
- dioxid
- transit
- zero
- net
Optional · uses your own API key
Second opinion from an LLM
The model reads this claim and the evidence passages shown on this page, picks one of the four verdicts and cites the passages it used. It is told to judge from the passages only. You review the answer; the call and your decision go to the AI audit log in this browser.
Without a key nothing is sent anywhere. The rest of the site does not need one.