SOTAVerified

Common Sense Reasoning

Common sense reasoning tasks are intended to require the model to go beyond pattern recognition. Instead, the model should use "common sense" or world knowledge to make inferences.

Papers

Showing 151200 of 939 papers

TitleStatusHype
pair2vec: Compositional Word-Pair Embeddings for Cross-Sentence InferenceCode1
ProtoQA: A Question Answering Dataset for Prototypical Common-Sense ReasoningCode1
NeuSyRE: Neuro-Symbolic Visual Understanding and Reasoning Framework based on Scene Graph EnrichmentCode1
Multi-Modal Grounded Planning and Efficient Replanning For Learning Embodied Agents with A Few ExamplesCode1
DialogSum: A Real-Life Scenario Dialogue Summarization DatasetCode1
MWP-BERT: Numeracy-Augmented Pre-training for Math Word Problem SolvingCode1
Evaluation Toolkit For Robustness Testing Of Automatic Essay Scoring SystemsCode1
Playing Codenames with Language Graphs and Word EmbeddingsCode1
ByteSized32: A Corpus and Challenge Task for Generating Task-Specific World Models Expressed as Text GamesCode1
Model Surgery: Modulating LLM's Behavior Via Simple Parameter EditingCode1
Does It Make Sense? And Why? A Pilot Study for Sense Making and ExplanationCode1
Event2Mind for Russian: Understanding Emotions and Intents in Texts. Corpus and Model for EvaluationCode1
mT5: A massively multilingual pre-trained text-to-text transformerCode1
Evidence-Aware Inferential Text Generation with Vector Quantised Variational AutoEncoderCode1
Multi-label Classification with High-rank and High-order Label CorrelationsCode1
Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense ReasoningCode1
Boosting Theory-of-Mind Performance in Large Language Models via PromptingCode1
Blow the Dog Whistle: A Chinese Dataset for Cant Understanding with Common Sense and World KnowledgeCode1
LLM-Coordination: Evaluating and Analyzing Multi-agent Coordination Abilities in Large Language ModelsCode1
Exploring the Benefits of Training Expert Language Models over Instruction TuningCode1
Memory Sharing for Large Language Model based AgentsCode1
Conversational Word Embedding for Retrieval-Based Dialog SystemCode1
Finding Effective Security Strategies through Reinforcement Learning and Self-PlayCode1
Fake News Detection on Social Media using Geometric Deep LearningCode1
Mitigating the Alignment Tax of RLHFCode1
Common Sense or World Knowledge? Investigating Adapter-Based Knowledge Injection into Pretrained TransformersCode1
Mention Flags (MF): Constraining Transformer-based Text GeneratorsCode1
Mobile App Tasks with Iterative Feedback (MoTIF): Addressing Task Feasibility in Interactive Visual EnvironmentsCode1
OpenBA-V2: Reaching 77.3% High Compression Ratio with Fast Multi-Stage PruningCode1
Generating similes effortlessly like a Pro: A Style Transfer Approach for Simile GenerationCode1
LoSiA: Efficient High-Rank Fine-Tuning via Subnet Localization and OptimizationCode0
Being Right for Whose Right Reasons?Code0
An Evaluation of PredPatt and Open IE via Stage 1 Semantic Role LabelingCode0
A Content-Based Novelty Measure for Scholarly Publications: A Proof of ConceptCode0
BAMO at SemEval-2024 Task 9: BRAINTEASER: A Novel Task Defying Common SenseCode0
A Neural Conversational ModelCode0
Learning Emphasis Selection for Written Text in Visual Media from Crowd-Sourced Label DistributionsCode0
Learning Low-Level Causal Relations using a Simulated Robotic ArmCode0
A Group-Specific Approach to NLP for Hate Speech DetectionCode0
Learn How to Cook a New Recipe in a New House: Using Map Familiarization, Curriculum Learning, and Bandit Feedback to Learn Families of Text-Based Adventure GamesCode0
Learning structure-aware semantic segmentation with image-level supervisionCode0
Large Language Models Need Consultants for Reasoning: Becoming an Expert in a Complex Human System Through Behavior SimulationCode0
An Analysis of Dataset Overlap on Winograd-Style TasksCode0
LVLM-Compress-Bench: Benchmarking the Broader Impact of Large Vision-Language Model CompressionCode0
Knowledge-Driven Robot Program Synthesis from Human VR DemonstrationsCode0
KnowZRel: Common Sense Knowledge-based Zero-Shot Relationship Retrieval for Generalised Scene Graph GenerationCode0
KC-ISA: An Implicit Sentiment Analysis Model Combining Knowledge Enhancement and Context FeaturesCode0
AmbiK: Dataset of Ambiguous Tasks in Kitchen EnvironmentCode0
“It doesn’t look good for a date”: Transforming Critiques into Preferences for Conversational Recommendation SystemsCode0
Leveraging QA Datasets to Improve Generative Data AugmentationCode0
Show:102550
← PrevPage 4 of 19Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1ST-MoE-32B 269B (fine-tuned)Accuracy96.1Unverified
2Unicorn 11B (fine-tuned)Accuracy91.3Unverified
3CompassMTL 567M with TailorAccuracy90.5Unverified
4CompassMTL 567MAccuracy89.6Unverified
5UnifiedQA 11B (fine-tuned)Accuracy89.4Unverified
6Claude 3 Opus (5-shot)Accuracy88.5Unverified
7GPT-4 (5-shot)Accuracy87.5Unverified
8ExDeBERTa 567MAccuracy87Unverified
9LLaMA-2 13B + MixLoRAAccuracy86.3Unverified
10LLaMA3 8B+MoSLoRAAccuracy85.8Unverified
#ModelMetricClaimedVerifiedStatus
1GPT-4 (few-shot, k=25)Accuracy96.4Unverified
2PaLM 2 (few-shot, CoT, SC)Accuracy95.1Unverified
3Shivaay (4B, few-shot, k=8)Accuracy91.04Unverified
4StupidLLMAccuracy91.03Unverified
5Claude 2 (few-shot, k=5)Accuracy91Unverified
6Claude 1.3 (few-shot, k=5)Accuracy90Unverified
7PaLM 540B (Self Improvement, Self Consistency)Accuracy89.8Unverified
8PaLM 540B (Self Consistency)Accuracy88.7Unverified
9PaLM 540B (Self Improvement, CoT Prompting)Accuracy88.3Unverified
10PaLM 540B (Self Improvement, Standard-Prompting)Accuracy87.2Unverified
#ModelMetricClaimedVerifiedStatus
1ST-MoE-32B 269B (fine-tuned)Accuracy95.2Unverified
2LLaMA 3 8B+MoSLoRA (fine-tuned)Accuracy90.5Unverified
3PaLM 2-L (1-shot)Accuracy89.7Unverified
4PaLM 2-M (1-shot)Accuracy88Unverified
5LLaMA-3 8B + MixLoRAAccuracy86.5Unverified
6Camelidae-8×34BAccuracy86.2Unverified
7PaLM 2-S (1-shot)Accuracy85.6Unverified
8LLaMA 65B + CFG (0-shot)Accuracy84.2Unverified
9GAL 120B (0-shot)Accuracy83.8Unverified
10LLaMA-2 13B + MixLoRAAccuracy83.5Unverified
#ModelMetricClaimedVerifiedStatus
1Turing NLR v5 XXL 5.4B (fine-tuned)EM95.9Unverified
2ST-MoE-32B 269B (fine-tuned)EM95.1Unverified
3T5-11BF194.1Unverified
4DeBERTa-1.5BEM94.1Unverified
5PaLM 540B (finetuned)EM94Unverified
6Vega v2 6B (fine-tuned)EM93.9Unverified
7PaLM 2-L (one-shot)F193.8Unverified
8T5-XXL 11B (fine-tuned)EM93.4Unverified
9PaLM 2-M (one-shot)F192.4Unverified
10PaLM 2-S (one-shot)F192.1Unverified