Merge branch 'main' into group-agg-rework

lm_eval/tasks/bertaqa/bertaqa_en_mt_latxa-13b-v1.yaml

+task: bertaqa_en_mt_latxa-13b-v1
+include: _bertaqa_template
+dataset_name: en_mt_latxa-13b-v1
+doc_to_text: "Question: {{question}}\nA: {{candidates[0]}}\nB: {{candidates[1]}}\nC: {{candidates[2]}}\nAnswer:"

lm_eval/tasks/bertaqa/bertaqa_en_mt_latxa-70b-v1.1.yaml

+task: bertaqa_en_mt_latxa-70b-v1.1
+include: _bertaqa_template
+dataset_name: en_mt_latxa-70b-v1.1
+doc_to_text: "Question: {{question}}\nA: {{candidates[0]}}\nB: {{candidates[1]}}\nC: {{candidates[2]}}\nAnswer:"

lm_eval/tasks/bertaqa/bertaqa_en_mt_latxa-70b-v1.yaml

+task: bertaqa_en_mt_latxa-70b-v1
+include: _bertaqa_template
+dataset_name: en_mt_latxa-70b-v1
+doc_to_text: "Question: {{question}}\nA: {{candidates[0]}}\nB: {{candidates[1]}}\nC: {{candidates[2]}}\nAnswer:"

lm_eval/tasks/bertaqa/bertaqa_en_mt_latxa-7b-v1.1.yaml

+task: bertaqa_en_mt_latxa-7b-v1.1
+include: _bertaqa_template
+dataset_name: en_mt_latxa-7b-v1.1
+doc_to_text: "Question: {{question}}\nA: {{candidates[0]}}\nB: {{candidates[1]}}\nC: {{candidates[2]}}\nAnswer:"

lm_eval/tasks/bertaqa/bertaqa_en_mt_latxa-7b-v1.yaml

+task: bertaqa_en_mt_latxa-7b-v1
+include: _bertaqa_template
+dataset_name: en_mt_latxa-7b-v1
+doc_to_text: "Question: {{question}}\nA: {{candidates[0]}}\nB: {{candidates[1]}}\nC: {{candidates[2]}}\nAnswer:"

lm_eval/tasks/bertaqa/bertaqa_en_mt_llama-2-13b.yaml

+task: bertaqa_en_mt_llama-2-13b
+include: _bertaqa_template
+dataset_name: en_mt_llama-2-13b
+doc_to_text: "Question: {{question}}\nA: {{candidates[0]}}\nB: {{candidates[1]}}\nC: {{candidates[2]}}\nAnswer:"

lm_eval/tasks/bertaqa/bertaqa_en_mt_llama-2-70b.yaml

+task: bertaqa_en_mt_llama-2-70b
+include: _bertaqa_template
+dataset_name: en_mt_llama-2-70b
+doc_to_text: "Question: {{question}}\nA: {{candidates[0]}}\nB: {{candidates[1]}}\nC: {{candidates[2]}}\nAnswer:"

lm_eval/tasks/bertaqa/bertaqa_en_mt_llama-2-7b.yaml

+task: bertaqa_en_mt_llama-2-7b
+include: _bertaqa_template
+dataset_name: en_mt_llama-2-7b
+doc_to_text: "Question: {{question}}\nA: {{candidates[0]}}\nB: {{candidates[1]}}\nC: {{candidates[2]}}\nAnswer:"

lm_eval/tasks/bertaqa/bertaqa_en_mt_madlad.yaml

+task: bertaqa_en_mt_madlad
+include: _bertaqa_template
+dataset_name: en_mt_madlad
+doc_to_text: "Question: {{question}}\nA: {{candidates[0]}}\nB: {{candidates[1]}}\nC: {{candidates[2]}}\nAnswer:"

lm_eval/tasks/bertaqa/bertaqa_en_mt_nllb.yaml

+task: bertaqa_en_mt_nllb
+include: _bertaqa_template
+dataset_name: en_mt_nllb
+doc_to_text: "Question: {{question}}\nA: {{candidates[0]}}\nB: {{candidates[1]}}\nC: {{candidates[2]}}\nAnswer:"

lm_eval/tasks/bertaqa/bertaqa_eu.yaml

+task: bertaqa_eu
+include: _bertaqa_template
+dataset_name: eu
+doc_to_text: "Galdera: {{question}}\nA: {{candidates[0]}}\nB: {{candidates[1]}}\nC: {{candidates[2]}}\nErantzuna:"

lm_eval/tasks/bigbench/push_bigbench_dataset.py

@@ -8,6 +8,7 @@ Requires the installation of
 `pip install "bigbench @ https://storage.googleapis.com/public_research_data/bigbench/bigbench-0.0.1.tar.gz"`
 and is included so that the bigbench dependency can be avoided.
 """
+
 import bigbench.api.util as bb_utils
 import datasets
 from tqdm import tqdm

lm_eval/tasks/ceval/_generate_configs.py

 """
 Take in a YAML, and output all other splits with this YAML
 """
+
 import argparse
 import os
 

lm_eval/tasks/cmmlu/_generate_configs.py

 """
 Take in a YAML, and output all other splits with this YAML
 """
+
 import argparse
 import os
 

lm_eval/tasks/commonsense_qa/README.md

+# Task-name
+
+### Paper
+
+Title: `COMMONSENSEQA: A Question Answering Challenge Targeting
+Commonsense Knowledge`
+
+Abstract: https://arxiv.org/pdf/1811.00937.pdf
+
+CommonsenseQA is a multiple-choice question answering dataset that requires different types of commonsense knowledge to predict the correct answers.
+It contains 12,102 questions with one correct answer and four distractor answers.
+
+Homepage: https://www.tau-nlp.org/commonsenseqa
+
+
+### Citation
+
+```
+@inproceedings{talmor-etal-2019-commonsenseqa,
+    title = "{C}ommonsense{QA}: A Question Answering Challenge Targeting Commonsense Knowledge",
+    author = "Talmor, Alon  and
+      Herzig, Jonathan  and
+      Lourie, Nicholas  and
+      Berant, Jonathan",
+    booktitle = "Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)",
+    month = jun,
+    year = "2019",
+    address = "Minneapolis, Minnesota",
+    publisher = "Association for Computational Linguistics",
+    url = "https://aclanthology.org/N19-1421",
+    doi = "10.18653/v1/N19-1421",
+    pages = "4149--4158",
+    archivePrefix = "arXiv",
+    eprint        = "1811.00937",
+    primaryClass  = "cs",
+}
+```
+
+### Groups and Tasks
+
+#### Groups
+
+* Not part of a group yet.
+
+#### Tasks
+
+* `commonsense_qa`: Represents the "random" split from the paper. Uses an MMLU-style prompt, as (presumably) used by Llama evaluations.
+
+### Checklist
+
+For adding novel benchmarks/datasets to the library:
+* [x] Is the task an existing benchmark in the literature?
+  * [x] Have you referenced the original paper that introduced the task?
+  * [x] If yes, does the original paper provide a reference implementation? If so, have you checked against the reference implementation and documented how to run such a test?
+
+
+If other tasks on this dataset are already supported:
+* [ ] Is the "Main" variant of this task clearly denoted?
+* [ ] Have you provided a short sentence in a README on what each new variant adds / evaluates?
+* [ ] Have you noted which, if any, published evaluation setups are matched by this variant?

lm_eval/tasks/commonsense_qa/default.yaml

+task: commonsense_qa
+dataset_path: tau/commonsense_qa
+training_split: train
+validation_split: validation
+output_type: multiple_choice
+doc_to_text: "Question: {{ question.strip() }}\nA. {{choices['text'][0]}}\nB. {{choices['text'][1]}}\nC. {{choices['text'][2]}}\nD. {{choices['text'][3]}}\nE. {{choices['text'][4]}}\nAnswer:"
+doc_to_target: answerKey
+doc_to_choice: ['A', 'B', 'C', 'D', 'E']
+metric_list:
+  - metric: acc
+    aggregation: mean
+    higher_is_better: true

lm_eval/tasks/csatqa/_generate_configs.py

 """
 Take in a YAML, and output all other splits with this YAML
 """
+
 import argparse
 import os
 

lm_eval/tasks/fda/task.py

-"""
-"""
 import re
 from typing import List
 

lm_eval/tasks/fld/README.md

@@ -38,18 +38,19 @@ Homepage: https://github.com/hitachi-nlp/FLD
 
 ### Groups and Tasks
 
-#### Groups
-
-* `fld`
-
-#### Tasks
-
 This release is the simplified version of FLD where a model is required to predict only an answer.
 This setting is described by "answer accuracy" in the original paper.
 
+#### Tasks in Group `fld`
 * `fld_default` is a basic task based on [FLD.v2](https://huggingface.co/datasets/hitachi-nlp/FLD.v2/viewer/star)
 * `fld_star`: is a more challenging version based on [FLD.v2-star](https://huggingface.co/datasets/hitachi-nlp/FLD.v2/viewer/star)
 
+#### Tasks in Group `fld_logical_formula`
+Further, we have "logical formula" versions of the benchmarks, which evaluate LLMs' pure logical reasoning capabilities within the domain of logical formulas, rather than natural language:
+* `fld_logical_formula_default`
+* `fld_logical_formula_fld_star`
+
+
 ### Checklist
 
 For adding novel benchmarks/datasets to the library:

lm_eval/tasks/fld/fld_logical_formula_default.yaml

+group:
+  - fld_logical_formula
+task: fld_logical_formula_default
+dataset_path: hitachi-nlp/FLD.v2
+dataset_name: default
+training_split: train
+validation_split: validation
+test_split: test
+doc_to_text: "Based on the provided facts ($context$), either prove or disprove the hypothesis or state that it is unknown. The facts and the hypothesis are written in logical formulas as follows: capital letters such as \"{A}\", \"{B}\", \"{AB}\" are predicates, small letters such as \"{a}\", \"{b}\", \"{ab}\" are constants, \"&\" is logical conjunction, \"v\" is logical disjunction, \"¬\" is negation, \"->\" is implication, \"(x)\" is \"for all x\", and \"(Ex)\" is \"for some x\". $hypothesis$ = {{hypothesis_formula}} ; $context$ = {{context_formula}} ; $proof$ = "
+doc_to_target: world_assump_label
+metric_list:
+  - metric: exact_match
+    aggregation: mean
+    higher_is_better: true
+filter_list:
+  - name: remove_whitespace
+    filter:
+      - function: remove_whitespace
+      - function: take_first
+metadata:
+  version: 2.0