61 stories tagged Reasoning · page 2 of 3
Researchers developed a way to train AI models to handle real-world problems with incomplete or ambiguous information. This could improve AI's ability to make practical decisions in everyday scenarios.
Researchers developed a system that helps AI models save effort by quickly deciding whether a problem needs deep thought. This could make AI tools faster and more efficient for everyday users.
A new study proposes Visual Graph Scaffolds, a method that uses graph structures to improve the reasoning of large language models (LLMs). Unlike prior approaches that treat graphs as external data sources, this technique integrates graphs directly into the model's reasoning process, inspired by how humans use mind maps to organize complex thoughts. The method showed significant improvements on multi-hop question answering tasks.
A new study reveals that AI models can sometimes hurt their own performance by overthinking. Researchers from ArXiv cs.AI found that once a large reasoning model (LRM) reaches the correct answer, additional reasoning does not improve the response and can lead to new errors. This challenges the common assumption that longer reasoning always yields better results.
Researchers created a new way to test AI's reasoning skills using interactive games. This method evaluates how well AI can gather information, adapt, and make decisions—just like solving a mystery.
Researchers introduced MindLoom, a framework that improves AI reasoning by breaking down complex problems into simpler thought modes. This could lead to more reliable and diverse problem-solving in AI systems.
Researchers developed a technique called ProxyCoT to help AI models reason better with long texts. This method trains models to use shorter, relevant parts of long documents to solve complex problems.
Scientists discovered that AI models trained for reasoning don't just think longer—they actually move differently in their internal processes. This changes how we understand and improve AI problem-solving. The study found that longer chains of thought don't necessarily mean better reasoning, but rather a different internal path. This could lead to more efficient and effective AI models in the future.
A new technique called verifiable process supervision (VPS) helps AI models produce both accurate answers and sound reasoning. This addresses a common problem where AI might guess correctly but use flawed logic.
Researchers introduced a new framework called CHAL for AI debates. It focuses on areas where truth is uncertain, helping AI models refine their reasoning through structured discussions. This could make AI more useful in real-world, ambiguous situations.
Researchers have developed a better way for AI to handle complex reasoning by tracking its thought process and knowing when to stop. This could make AI assistants and decision-making tools more reliable for everyday use.
Researchers discovered that longer reasoning processes in AI models can make them more biased. This challenges the assumption that more 'thinking' always leads to better, fairer results.
Researchers have developed BALAR, a system that helps AI models actively seek missing information in conversations. This could make AI assistants much more effective at multi-step tasks.
Researchers have developed a new reinforcement learning technique called Adaptive Power-Mean Policy Optimization (APMPO) that improves how AI models reason. This method adapts to the evolving capabilities of large language models, making them more effective at problem-solving.
Researchers found that deep AI models can perform deductive reasoning nearly as well as models that follow step-by-step logic. This could make AI smarter without needing extra instructions.
Researchers developed DIAGRAMS, a tool to help AI explain its reasoning when answering questions about diagrams. This makes it easier to understand how AI arrives at its answers, improving transparency.
A new study explores whether fundamental reasoning patterns in LLMs can be decoupled from specific problem instances. The research highlights the challenges and implications for model controllability and reasoning capabilities.
A new study questions the assumption that symbol grounding automatically leads to compositional reasoning in neuro-symbolic systems. The research introduces a novel framework to disentangle these two critical capabilities.
Researchers discovered that providing one confirmed fact in a multi-step reasoning chain increases LLMs' likelihood of confidently producing wrong answers. This phenomenon, called anchored confabulation, challenges assumptions about how models handle partial evidence.
A new study identifies flaws in how LLMs handle clinical trial data, proposing a hybrid approach to improve reasoning. The research focuses on recovering implicit attributes from partially observed tables.
Researchers introduce a new benchmark, Math Takes Two, to evaluate whether language models truly understand math or just memorize patterns. The test focuses on emergent mathematical reasoning through communication, challenging models to construct abstract concepts from first principles.
Researchers introduce TRACES, a method to tag and analyze reasoning steps in Language Reasoning Models (LRMs). This approach aims to reduce inefficiencies and improve the accuracy of model outputs.
A new position paper challenges the prevailing view of LLM reasoning as a chain of thought, proposing instead that it should be studied as latent-state trajectory formation. This shift could redefine how we evaluate and interpret AI reasoning.
Researchers introduce LACE, a framework that allows parallel reasoning paths in LLMs to interact and correct each other. This could significantly improve the robustness of model outputs.