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

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–50 of 939 papers

TitleStatusHype
Comparing Apples to Oranges: A Dataset & Analysis of LLM Humour Understanding from Traditional Puns to Topical Jokes—0
LoSiA: Efficient High-Rank Fine-Tuning via Subnet Localization and OptimizationCode0
CheckManual: A New Challenge and Benchmark for Manual-based Appliance Manipulation—0
EditInspector: A Benchmark for Evaluation of Text-Guided Image Edits—0
Prime the search: Using large language models for guiding geometric task and motion planning by warm-starting tree searchCode0
AmbiK: Dataset of Ambiguous Tasks in Kitchen EnvironmentCode0
ATLAS: Learning to Optimally Memorize the Context at Test Time—0
Spatial Knowledge Graph-Guided Multimodal Synthesis—0
CaseEdit: Enhancing Localized Commonsense Reasoning via Null-Space Constrained Knowledge Editing in Small Parameter Language Models—0
SOLVE: Synergy of Language-Vision and End-to-End Networks for Autonomous Driving—0
Align-GRAG: Reasoning-Guided Dual Alignment for Graph Retrieval-Augmented Generation—0
OSoRA: Output-Dimension and Singular-Value Initialized Low-Rank Adaptation—0
3D Visual Illusion Depth EstimationCode1
Empirically evaluating commonsense intelligence in large language models with large-scale human judgments—0
ProdRev: A DNN framework for empowering customers using generative pre-trained transformers—0
Through the Looking Glass: Common Sense Consistency Evaluation of Weird Images—0
AgentSGEN: Multi-Agent LLM in the Loop for Semantic Collaboration and GENeration of Synthetic Data—0
Scenethesis: A Language and Vision Agentic Framework for 3D Scene Generation—0
VideoHallu: Evaluating and Mitigating Multi-modal Hallucinations on Synthetic Video UnderstandingCode1
UAV-VLN: End-to-End Vision Language guided Navigation for UAVs—0
ScanEdit: Hierarchically-Guided Functional 3D Scan Editing—0
CheXWorld: Exploring Image World Modeling for Radiograph Representation LearningCode1
Creating 'Full-Stack' Hybrid Reasoning Systems that Prioritize and Enhance Human Intelligence—0
FLIP Reasoning ChallengeCode0
Shrinkage Initialization for Smooth Learning of Neural Networks—0
JEPA4Rec: Learning Effective Language Representations for Sequential Recommendation via Joint Embedding Predictive Architecture—0
What the HellaSwag? On the Validity of Common-Sense Reasoning BenchmarksCode0
InstructionBench: An Instructional Video Understanding Benchmark—0
Proposition of Affordance-Driven Environment Recognition Framework Using Symbol Networks in Large Language Models—0
DynMoLE: Boosting Mixture of LoRA Experts Fine-Tuning with a Hybrid Routing MechanismCode0
WinoWhat: A Parallel Corpus of Paraphrased WinoGrande Sentences with Common Sense Categorization—0
Information Gain Is Not All You NeedCode0
GLRD: Global-Local Collaborative Reason and Debate with PSL for 3D Open-Vocabulary Detection—0
Global-Local Tree Search in VLMs for 3D Indoor Scene GenerationCode1
Unbiasing through Textual Descriptions: Mitigating Representation Bias in Video Benchmarks—0
A Study on Neuro-Symbolic Artificial Intelligence: Healthcare Perspectives—0
Improving Preference Extraction In LLMs By Identifying Latent Knowledge Through Classifying Probes—0
Don't Fight Hallucinations, Use Them: Estimating Image Realism using NLI over Atomic FactsCode0
Cosmos-Reason1: From Physical Common Sense To Embodied ReasoningCode4
Do I look like a `cat.n.01` to you? A Taxonomy Image Generation Benchmark—0
HybridVLA: Collaborative Diffusion and Autoregression in a Unified Vision-Language-Action Model—0
AlphaDrive: Unleashing the Power of VLMs in Autonomous Driving via Reinforcement Learning and ReasoningCode3
WISE: A World Knowledge-Informed Semantic Evaluation for Text-to-Image GenerationCode4
LVLM-Compress-Bench: Benchmarking the Broader Impact of Large Vision-Language Model CompressionCode0
The Box is in the Pen: Evaluating Commonsense Reasoning in Neural Machine TranslationCode0
LLM-Advisor: An LLM Benchmark for Cost-efficient Path Planning across Multiple Terrains—0
Code-as-Symbolic-Planner: Foundation Model-Based Robot Planning via Symbolic Code Generation—0
Personalized Causal Graph Reasoning for LLMs: A Case Study on Dietary Recommendations—0
FRIDA to the Rescue! Analyzing Synthetic Data Effectiveness in Object-Based Common Sense Reasoning for Disaster Response—0
The Lottery LLM Hypothesis, Rethinking What Abilities Should LLM Compression Preserve?—0
Show:102550
← PrevPage 1 of 19Next →
All datasetsWinoGrandearc_challengearc_easyReCoRDCommonsenseQAPARusRuCoSRWSDBIG-bench (Causal Judgment)BIG-bench (Date Understanding)BIG-bench (Disambiguation QA)BIG-bench (Sports Understanding)

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1ST-MoE-32B 269B (fine-tuned)Accuracy96.1—Unverified
2Unicorn 11B (fine-tuned)Accuracy91.3—Unverified
3CompassMTL 567M with TailorAccuracy90.5—Unverified
4CompassMTL 567MAccuracy89.6—Unverified
5UnifiedQA 11B (fine-tuned)Accuracy89.4—Unverified
6Claude 3 Opus (5-shot)Accuracy88.5—Unverified
7GPT-4 (5-shot)Accuracy87.5—Unverified
8ExDeBERTa 567MAccuracy87—Unverified
9LLaMA-2 13B + MixLoRAAccuracy86.3—Unverified
10LLaMA3 8B+MoSLoRAAccuracy85.8—Unverified
#ModelMetricClaimedVerifiedStatus
1GPT-4 (few-shot, k=25)Accuracy96.4—Unverified
2PaLM 2 (few-shot, CoT, SC)Accuracy95.1—Unverified
3Shivaay (4B, few-shot, k=8)Accuracy91.04—Unverified
4StupidLLMAccuracy91.03—Unverified
5Claude 2 (few-shot, k=5)Accuracy91—Unverified
6Claude 1.3 (few-shot, k=5)Accuracy90—Unverified
7PaLM 540B (Self Improvement, Self Consistency)Accuracy89.8—Unverified
8PaLM 540B (Self Consistency)Accuracy88.7—Unverified
9PaLM 540B (Self Improvement, CoT Prompting)Accuracy88.3—Unverified
10PaLM 540B (Self Improvement, Standard-Prompting)Accuracy87.2—Unverified
#ModelMetricClaimedVerifiedStatus
1ST-MoE-32B 269B (fine-tuned)Accuracy95.2—Unverified
2LLaMA 3 8B+MoSLoRA (fine-tuned)Accuracy90.5—Unverified
3PaLM 2-L (1-shot)Accuracy89.7—Unverified
4PaLM 2-M (1-shot)Accuracy88—Unverified
5LLaMA-3 8B + MixLoRAAccuracy86.5—Unverified
6Camelidae-8×34BAccuracy86.2—Unverified
7PaLM 2-S (1-shot)Accuracy85.6—Unverified
8LLaMA 65B + CFG (0-shot)Accuracy84.2—Unverified
9GAL 120B (0-shot)Accuracy83.8—Unverified
10LLaMA-2 13B + MixLoRAAccuracy83.5—Unverified
#ModelMetricClaimedVerifiedStatus
1Turing NLR v5 XXL 5.4B (fine-tuned)EM95.9—Unverified
2ST-MoE-32B 269B (fine-tuned)EM95.1—Unverified
3T5-11BF194.1—Unverified
4DeBERTa-1.5BEM94.1—Unverified
5PaLM 540B (finetuned)EM94—Unverified
6Vega v2 6B (fine-tuned)EM93.9—Unverified
7PaLM 2-L (one-shot)F193.8—Unverified
8T5-XXL 11B (fine-tuned)EM93.4—Unverified
9PaLM 2-M (one-shot)F192.4—Unverified
10PaLM 2-S (one-shot)F192.1—Unverified