WildTableBench is the first QA benchmark for naturally occurring table images, where 21 multimodal models were evaluated and only one exceeded 50% accuracy.
hub Baseline reference
arXiv preprint arXiv:2209.14610 , year =
Baseline reference. 67% of citing Pith papers use this work as a benchmark or comparison.
20
Pith papers citing it
Baseline
67% of classified citations
abstract
Mathematical reasoning, a core ability of human intelligence, presents unique challenges for machines in abstract thinking and logical reasoning. Recent large pre-trained language models such as GPT-3 have achieved remarkable progress on mathematical reasoning tasks written in text form, such as math word problems (MWP). However, it is unknown if the models can handle more complex problems that involve math reasoning over heterogeneous information, such as tabular data. To fill the gap, we present Tabular Math Word Problems (TabMWP), a new dataset containing 38,431 open-domain grade-level problems that require mathematical reasoning on both textual and tabular data. Each question in TabMWP is aligned with a tabular context, which is presented as an image, semi-structured text, and a structured table. There are two types of questions: free-text and multi-choice, and each problem is annotated with gold solutions to reveal the multi-step reasoning process. We evaluate different pre-trained models on TabMWP, including the GPT-3 model in a few-shot setting. As earlier studies suggest, since few-shot GPT-3 relies on the selection of in-context examples, its performance is unstable and can degrade to near chance. The unstable issue is more severe when handling complex problems like TabMWP. To mitigate this, we further propose a novel approach, PromptPG, which utilizes policy gradient to learn to select in-context examples from a small amount of training data and then constructs the corresponding prompt for the test example. Experimental results show that our method outperforms the best baseline by 5.31% on the accuracy metric and reduces the prediction variance significantly compared to random selection, which verifies its effectiveness in selecting in-context examples.
hub tools
citation-role summary
dataset 4
background 2
citation-polarity summary
representative citing papers
Cross-modal agreement between chain-of-thought and program-of-thought reasoning enables self-consistency with only two LLM samples, reducing sampling cost by 9.3x while improving accuracy.
SARL optimizes language prompt inputs to generalist vision-language-action policies through online RL to solve complex long-horizon tasks by composing existing skills.
An automatic numeric-remapping attack generator reveals 12-26 point accuracy drops on GSM8K for three LLMs while MAWPS and MultiArith stay near 98%.
FLARE is a vision-language model family using text-guided vision encoding, context-aware alignment decoding, dual-semantic mapping loss, and text-driven VQA synthesis to achieve deep cross-modal integration, outperforming larger models with only 630 vision tokens at 3B scale.
PoT prompting improves numerical reasoning by having language models write programs executed by a computer instead of performing calculations in natural language chains of thought, with an average 12% gain over CoT.
Generalizes contrastive decoding to a conflict-aware paradigm identifying a power family with regime asymmetry, introduces TriState-Bench and Adaptive Regime Routing to improve resistance to erroneous context.
Zamba2-VL is a family of 1.2B–7B hybrid Mamba2-transformer vision-language models that match leading transformer VLMs on image, reasoning, OCR, grounding and counting benchmarks while delivering roughly 10x lower time-to-first-token.
V-tableR1 uses a critic VLM for dense step-level feedback and a new PGPO algorithm to shift multimodal table reasoning from pattern matching to verifiable logical steps, achieving SOTA accuracy with a 4B open-source model.
FSLR explicitly supervises the initial logical planning step in math problems, boosting LLM accuracy by 3-5% while using 80% fewer training tokens than standard CoT fine-tuning.
A two-stage RL framework first boosts text reasoning in 3B LMMs then adapts it to multimodal inputs, producing modest benchmark gains of 4.5-4.8%.
InternVL 2.5 is the first open-source MLLM to surpass 70% on the MMMU benchmark via model, data, and test-time scaling, with a 3.7-point gain from chain-of-thought reasoning.
Multimodal-CoT achieves state-of-the-art on ScienceQA by using a two-stage process that incorporates vision into chain-of-thought rationale generation for models under 1 billion parameters.
MoCA-Agent decomposes questions into typed atomic claims, clears them via trader-agent markets into confidence-weighted decisions, synthesizes and verifies executable Python code, and reports strong benchmark scores including 78.3% on FinQA.
Closed-loop self-evolution on LLMs improves reasoning on Knights and Knaves tasks but plateaus short of oracle-supervised levels, with multi-turn revision nearly matching it for large models.
InternLM-XComposer-2.5 is a 7B vision-language model supporting up to 96K context that reaches GPT-4V-level performance on image, video, and multi-turn tasks and adds LoRA-driven text-image composition capabilities.
ZAYA1-VL-8B is a new MoE vision-language model with vision-specific LoRA adapters and bidirectional image attention that reports competitive performance against several 3B-4B models on image, reasoning, and counting benchmarks.
InternVL 1.5 narrows the performance gap to proprietary multimodal models via a stronger transferable vision encoder, dynamic high-resolution tiling, and curated English-Chinese training data.
GPT-4V processes interleaved image-text inputs generically and supports visual referring prompting for new human-AI interaction.
A survey that categorizes TQA benchmarks and LLM modeling strategies by challenges while identifying underexplored areas such as reinforcement learning.
citing papers explorer
-
WildTableBench: Benchmarking Multimodal Foundation Models on Table Understanding In the Wild
WildTableBench is the first QA benchmark for naturally occurring table images, where 21 multimodal models were evaluated and only one exceeded 50% accuracy.
-
Self-Consistency from Only Two Samples: CoT-PoT Ensembling for Efficient LLM Reasoning
Cross-modal agreement between chain-of-thought and program-of-thought reasoning enables self-consistency with only two LLM samples, reducing sampling cost by 9.3x while improving accuracy.
-
Adapting Generalist Robot Policies with Semantic Reinforcement Learning
SARL optimizes language prompt inputs to generalist vision-language-action policies through online RL to solve complex long-horizon tasks by composing existing skills.
-
Testing LLM Arithmetic Reasoning Generalization with Automatic Numeric-Remapping Attacks
An automatic numeric-remapping attack generator reveals 12-26 point accuracy drops on GSM8K for three LLMs while MAWPS and MultiArith stay near 98%.
-
FLARE: Fully Integration of Vision-Language Representations for Deep Cross-Modal Understanding
FLARE is a vision-language model family using text-guided vision encoding, context-aware alignment decoding, dual-semantic mapping loss, and text-driven VQA synthesis to achieve deep cross-modal integration, outperforming larger models with only 630 vision tokens at 3B scale.
-
Program of Thoughts Prompting: Disentangling Computation from Reasoning for Numerical Reasoning Tasks
PoT prompting improves numerical reasoning by having language models write programs executed by a computer instead of performing calculations in natural language chains of thought, with an average 12% gain over CoT.
-
From Context-Aware to Conflict-Aware: Generalizing Contrastive Decoding for Knowledge Conflict in LLMs
Generalizes contrastive decoding to a conflict-aware paradigm identifying a power family with regime asymmetry, introduces TriState-Bench and Adaptive Regime Routing to improve resistance to erroneous context.
-
Zamba2-VL Technical Report
Zamba2-VL is a family of 1.2B–7B hybrid Mamba2-transformer vision-language models that match leading transformer VLMs on image, reasoning, OCR, grounding and counting benchmarks while delivering roughly 10x lower time-to-first-token.
-
V-tableR1: Process-Supervised Multimodal Table Reasoning with Critic-Guided Policy Optimization
V-tableR1 uses a critic VLM for dense step-level feedback and a new PGPO algorithm to shift multimodal table reasoning from pattern matching to verifiable logical steps, achieving SOTA accuracy with a 4B open-source model.
-
From Implicit to Explicit: Token-Efficient Logical Supervision for Mathematical Reasoning in LLMs
FSLR explicitly supervises the initial logical planning step in math problems, boosting LLM accuracy by 3-5% while using 80% fewer training tokens than standard CoT fine-tuning.
-
LMM-R1: Empowering 3B LMMs with Strong Reasoning Abilities Through Two-Stage Rule-Based RL
A two-stage RL framework first boosts text reasoning in 3B LMMs then adapts it to multimodal inputs, producing modest benchmark gains of 4.5-4.8%.
-
Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling
InternVL 2.5 is the first open-source MLLM to surpass 70% on the MMMU benchmark via model, data, and test-time scaling, with a 3.7-point gain from chain-of-thought reasoning.
-
Multimodal Chain-of-Thought Reasoning in Language Models
Multimodal-CoT achieves state-of-the-art on ScienceQA by using a two-stage process that incorporates vision into chain-of-thought rationale generation for models under 1 billion parameters.
-
MoCA-Agent: A Market-of-Claims Code Agent for Financial and Numerical Reasoning
MoCA-Agent decomposes questions into typed atomic claims, clears them via trader-agent markets into confidence-weighted decisions, synthesizes and verifies executable Python code, and reports strong benchmark scores including 78.3% on FinQA.
-
On the Generalization Gap in Self-Evolving Language Model Reasoning
Closed-loop self-evolution on LLMs improves reasoning on Knights and Knaves tasks but plateaus short of oracle-supervised levels, with multi-turn revision nearly matching it for large models.
-
InternLM-XComposer-2.5: A Versatile Large Vision Language Model Supporting Long-Contextual Input and Output
InternLM-XComposer-2.5 is a 7B vision-language model supporting up to 96K context that reaches GPT-4V-level performance on image, video, and multi-turn tasks and adds LoRA-driven text-image composition capabilities.
-
ZAYA1-VL-8B Technical Report
ZAYA1-VL-8B is a new MoE vision-language model with vision-specific LoRA adapters and bidirectional image attention that reports competitive performance against several 3B-4B models on image, reasoning, and counting benchmarks.
-
How Far Are We to GPT-4V? Closing the Gap to Commercial Multimodal Models with Open-Source Suites
InternVL 1.5 narrows the performance gap to proprietary multimodal models via a stronger transferable vision encoder, dynamic high-resolution tiling, and curated English-Chinese training data.
-
The Dawn of LMMs: Preliminary Explorations with GPT-4V(ision)
GPT-4V processes interleaved image-text inputs generically and supports visual referring prompting for new human-AI interaction.
-
Table Question Answering in the Era of Large Language Models: A Comprehensive Survey of Tasks, Methods, and Evaluation
A survey that categorizes TQA benchmarks and LLM modeling strategies by challenges while identifying underexplored areas such as reinforcement learning.