arXiv AI
Jul 21
Quantizing Recursive Reasoning Models
arXiv:2607. 16237v1 Announce Type: cross Abstract: Recursive reasoning models solve hard puzzles by applying compact, weight-tied blocks over many refinement steps.
From Fuzzy to Exact: The Halo Architecture for Infinite-Depth Reasoning via Rational Arithmetic
Related stories
arXiv AI
Aug 14
Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement
arXiv:2608. 13129v1 Announce Type: new Abstract: Large language models (LLMs) achieve strong results on mathematical reasoning benchmarks yet remain unreliable on elementary numerical tasks, including magnitude comparison, large-integer arithmetic, fractions, and scientific notation.
arXiv AI
Jun 2
eMoT: evolving Memory-of-Thought via Symbolic Anchoring and Memory Corrosion
arXiv:2606. 02054v1 Announce Type: new Abstract: While Large Language Models (LLMs) achieve impressive performance on multi-step reasoning tasks, their reliability is persistently hindered by critical limitations such as unconstrained hallucinations and poor numerical computation.
arXiv Machine Learning
Oct 8
Algorithmic Scratchpads and Curriculum Staging for Arithmetic Reasoning in Tiny Transformers
arXiv:2610.09003v1 Announce Type: new Abstract: Autoregressive Large Language Models (LLMs) frequently struggle with deterministic multi-step algorithmic tasks such as multi-digit multiplication and...
Hugging Face Trending Papers
Aug 13
Numeracy in Large Language Models: Fundamental Limitations and Paths to Improvement
Large language models (LLMs) achieve strong results on mathematical reasoning benchmarks yet remain unreliable on elementary numerical tasks, including magnitude comparison, large-integer arithmetic, fractions, and scientific notation. This survey examines basic numerical understanding as a capability distinct from high-level mathematical reasoning.
arXiv Machine Learning
Jul 28
When Can Depth Replace Precision? A Resource Theory of Quantized Neural Computation
arXiv:2607. 23390v1 Announce Type: new Abstract: When can additional low-bit residual computation replace missing numerical precision for a fixed input-output map?
arXiv Machine Learning
Oct 15
Rational Transductors
arXiv:2602.07599v2 Announce Type: replace Abstract: Standard Transformers excel at semantic modeling but struggle with rigid sequential logic and state tracking. Theoretical work establishes that...
arXiv AI
Oct 16
Fractional Power-of-Two Quantization for Efficient and Effective Multiplier-Free LLM Inference
The paper introduces Fractional Power-of-Two (FPoT) quantization, a base‑√2 weight grid that enables 4‑bit weight quantization for large language models while keeping a fully integer shift‑add datapath. FPoT combines a fractional PoT grid with uniform integer activations, a calibration‑time grid‑alignment refinement, and a lightweight 1‑adder √2 approximation to reduce gate count compared to a 4‑bit multiplier. On Llama‑3‑8B, FPoT boosts zero‑shot commonsense reasoning by 8.72% over the standard PoT baseline and stays within 0.4% of uniform integer performance, with benefits extending across seven models and three bit‑widths.
arXiv AI
Sep 15
Tool Use Reduces Depth-Induced Collapse in OOD Reasoning
arXiv:2602.21061v2 Announce Type: replace Abstract: Many current paths to more advanced AI depend on the assumption that large language models (LLMs) can generalize learned relationships to solve com...
Hugging Face Trending Papers
Jun 1
Extreme Low-Bit Inference in Reasoning Models: Failure Modes and Targeted Recovery
Large Reasoning Models (LRMs) rely on long reasoning traces, making inference expensive. While low-bit quantization reduces per-token decoding cost, we show that aggressive 2-bit inference can fail to deliver end-to-end speedup because instability in the generation process inflates total token count.
arXiv AI
Jun 10
ASyMOB: Algebraic Symbolic Mathematical Operations Benchmark
arXiv:2505. 23851v2 Announce Type: replace-cross Abstract: Large language models (LLMs) are increasingly applied to symbolic mathematics, yet existing evaluations often conflate pattern memorization with genuine reasoning.
arXiv AI
Jun 2
Extreme Low-Bit Inference in Reasoning Models: Failure Modes and Targeted Recovery
arXiv:2606. 02011v1 Announce Type: new Abstract: Large Reasoning Models (LRMs) rely on long reasoning traces, making inference expensive.