Search papers, labs, and topics across Lattice.
9 papers from Amazon Science on Reasoning & Chain-of-Thought
LLMs can achieve better zero-shot product ranking with 57% less token usage by reasoning over structured attribute graphs instead of raw text.
Targeted neuro-symbolic integration can reduce content bias in syllogistic reasoning, achieving over 94% accuracy while cutting content effects by 16%.
Domain-specific fine-tuning can induce "agentic collapse" in LLMs, but a surprisingly small amount of agentic data from *another* domain can bring those general tool-use skills roaring back.
Memory-augmented LLMs get a strategic upgrade: MemMA uses multi-agent reasoning to proactively guide memory construction and repair, leading to significant performance gains.
LLM reasoning research is inadvertently paving a dangerous path towards AI situational awareness and strategic deception, demanding a re-evaluation of current safety measures.
Save 20% on LLM costs with <2% accuracy drop by strategically cascading a small model with a large one, guided by a confidence-calibrated SLM.
LLMs can ace math problems while reasoning like a drunk toddler, with 82% of correct answers arising from unstable, inconsistent logic.
Latent reasoning models often take shortcuts to achieve high accuracy, and stronger supervision, while mitigating this, paradoxically restricts the diversity of their latent representations.
Forget fine-tuning: inject targeted time-series insights into general LLMs and watch their reasoning skills skyrocket by up to 26%.
