LLMs & Models
Large language models: GPT, Claude, Gemini, Mistral and open weights.
93 published articles
AI Research
Penelope hides its reasoning in a single decoder layer to cut inference costs
Penelope, a latent-reasoning framework for decoder-only Transformers, localizes recurrent computation to a narrow decoder interval, cutting inference latency without a major accuracy hit. The paper describes a curriculum that shifts reasoning from visible tokens to an internal GRU loop.
2026-08-04
AI Research
The monitor that goes silent when AI reasoning fails
Novelis Research shows token log-probability fails as a decoder monitor for quantized reasoning models, being blind to confident loops. They introduce a calibrated e-CUSUM controller that combines uncertainty and repetition signals, achieving selectivity on GSM8K with DeepSeek-R1-Distill-Qwen-1.5B.
2026-08-03
AI Research
The LLM that outsmarted every fixed formula for JD.com warehouse allocation
An LLM trained via solver-guided reinforcement learning picks the best MIP formulation for each inventory allocation instance at JD.com. Hit Ratio@1 jumped from 21% to 50%, and realized allocation accuracy beat every fixed formulation by 12.57 percentage points.
2026-08-03
K12-Bench: New Test of AI Curriculum Understanding
Why your AI tutor can't see how math builds on itself
Peking University's K12-Bench reveals that even the best language models barely understand how school concepts connect. Scores of 57% and 46% show a blind spot in AI's ability to handle prerequisite chains, concept taxonomies, and visual grounding, skills that real tutors use every day.
2026-08-02
Artificial Intelligence
Research idea generation gets 3.89x better with IDEAgent
IDEAgent uses lineages, multi-objective feedback, and sequential memory to balance quality and diversity in LLM-generated research ideas. Tested across 32 topics in 8 CS domains, it achieves 3.89x the Yield (diverse ideas above a quality threshold) of prior methods.
2026-08-02
3D Reasoning
SceneActBench: Why even the best VLMs fail at 3D action
SceneActBench tests eleven VLM configurations on five 3D tasks in a unified agent loop. Overall scores range from 38.6 to 50.2, with no model performing consistently. The benchmark exposes a blind spot in vision-language agents: acting on full scenes, not just describing them.
2026-07-31
Research Digest
The simple topology trick that beats prompt magic in LLM induction tasks
A new methodology called Hourglass reasoning enforces strict context isolation between induction, deduction, and implementation stages, passing only a compressed symbolic rule between them. On ARC-AGI-2, it raises best-of-5 accuracy by up to 14 points over iterative refinement; on ChipBench Verilog synthesis, it nearly doubles accuracy with GPT-5.5. Ablations confirm the topology itself drives the improvement.
2026-07-31
AI research
Over half your AI's reasoning is froth, and nobody noticed until now
LLMs often generate reasoning chains that are correct but padded with unnecessary steps. A new diagnostic benchmark shows current evaluators miss this inefficiency entirely, and half of human-written reasoning steps may be compressible.
2026-07-31
LLM distillation
LLM distillation stalls when context goes static. Flux-OPD makes it evolve
Contexts can carry task preferences into LLM training, but once distilled into a student they add little supervision. Peking University's Flux-OPD keeps the context moving with the student and uses a conflict term to weight teacher corrections. The abstract reports gains over existing OPD paradigms without naming numbers.
2026-07-31
AI Research
A 4.5 point score jump on MATH-500 from a monitoring controller that catches wandering models
New research proposes an external monitoring controller for quantized small language models that detects repetitive or degenerating reasoning paths and triggers a rollback and constrained re-decoding. Accuracy improved by 4.5 percentage points on a broad evaluation set, but the authors stress the findings are not confirmatory.
2026-07-30
Game AI
MAGIC solved the three problems that kept AI from building real game worlds
MAGIC is a four-stage pipeline that converts a single prompt into a playable Unity project with multiple connected scenes. It enforces portal reachability with a flood-fill validator and introduces an evaluation agent that actually walks through each transition, decades of metrics that only looked at single interiors.
2026-07-30
MemTensor's Memory Foundation Models
AI agents kept memory outside the model. Metis puts it inside
MemTensor's Metis is the first prototype of memory foundation models: history compressed into the backbone, read back through memory attention, updated in one forward pass. It challenges the vector-store era of agent memory. The limits are acknowledged but not yet quantified.
2026-07-30