Research Paper Cockpit

Daily Digest - 2026-11-12

Papers first seen in this daily snapshot.

Daily Archives

Quick jump into generated daily digests.

Research Workflow

Latest digest: 2026-11-17.

Papers

17 visible entries

huggingface Score 21.0

Incentivizing Self-Learning in Large Language Models via Reinforcement Learning

2026-10-29 · Lingkai Kong, Yuzhe Gu, Haiteng Zhao, Zijian Wu, Youqing Fang, Zhouqi Hua, Yanan Sun, Kai Chen, Wenwei Zhang

General AI

Large language models (LLMs) draw on parametric knowledge and retrieval-augmented generation (RAG) to solve knowledge-intensive tasks. Yet problem solving in unfamiliar domains calls for tightly coupled knowledge acquisition and reasoning, beyond factual retrieval and aggregation or learning solely from a preassembled …

Review
pending
Role
unreviewed
Read
now
huggingface Score 18.4

GTAGym: Real AAA Game Tasks Environment for Evaluation and Reinforcement Learning

2026-11-09 · Jiajun Chen, Shicheng Fang, Jiahao Lu, Yuxin Wang, Xipeng Qiu

General AI

AAA games offer a rich testbed for evaluating agent perception, reasoning, and control. However, evaluating whether agents can complete tasks in AAA games is challenging and often human-annotated because task outcomes are often difficult to verify automatically, let alone applying it for Reinforcement Learning (RL). We…

Review
pending
Role
unreviewed
Read
now
huggingface Score 11.4

CARTBench: Rubric-Based Evaluation of Shopping Advice for Complex and Ambiguous Requests with Timelines

2026-11-04 · Yuqin Dai, Zhanwei Zhang, Honghui Yang, Zhouheng Yao, Yueqi Zhang, Jingyu He, Chaoyang Shi, Fanding Wei, Boyuan Pan, Kai Xu, Yao Hu, Yongliang Shen

General AI

Large language model (LLM) shopping assistants now recommend products in multi-turn conversations. However, reliable evaluation faces Data scarcity and Difficulty Standardizing Criteria. To address these obstacles, we introduce CARTBench for CART (Complex and Ambiguous Requests with Timelines). For Data scarcity, we cu…

Review
pending
Role
unreviewed
Read
soon
huggingface Score 11.4

Ensemble Routing for Prompt-Adaptive Cost-Aware LLM Inference

2026-11-06 · Saul Urso, Emanuele Carlini, Patrizio Dazzi, Matteo Mordacchini

General AI

Large language models (LLMs) increasingly support Web-based services that must balance response quality and inference cost across heterogeneous requests. Conventional routing selects one model per prompt, while fixed ensembles cannot adapt their composition and cost to individual requests. We introduce PISA, a prompt-a…

Review
pending
Role
unreviewed
Read
now
huggingface Score 11.0

Scaling Mathematical Proof Supervision from Structured Textbooks

2026-10-29 · Youqing Fang, Zhouqi Hua, Yinhao Tang, Yuzhe Gu, Lingkai Kong, Bin Liu, Weiming Zhang, Kai Chen, Yanan Sun, Wenwei Zhang

General AI

Mathematical supervision for language models is typically flattened into isolated problem-solution pairs, discarding the curricular organization and evidential structure of the original sources. We introduce TOME-Math (Textbook-Oriented Mathematical Evidence), a large-scale corpus that converts 2,118 textbooks spanning…

Review
pending
Role
unreviewed
Read
soon
huggingface Score 10.0

Mixed Signals: Weight-Space Programming for EEG and ECG

2026-11-10 · Dara Varam, Taraneh Aminosharieh Najafi, David Atienza

General AI

Weight-space programming allows language models to read a trained neural network's weights and edit them. This opens the door to ``training'' by reasoning rather than through gradient descent. Prior work has demonstrated its functionality and promise for game-playing networks and fixing specific failures in language mo…

Review
pending
Role
unreviewed
Read
soon
huggingface Score 10.0

SteerDistill: Quantizing LLM Agents via Closed-Loop Teacher Steering

2026-11-10 · Genming Cui, Xiangyu Li, Zixu Hao, Xin Ding, Yunxin Liu, Ting Cao

General AI

Low-bit quantization reduces the inference cost of LLM agents but can compromise task success. Examining failed rollouts, we find that quantized agents overlook task requirements or environmental conditions, producing environment transitions that diverge from successful full-precision references. Correcting these behav…

Review
pending
Role
unreviewed
Read
soon
huggingface Score 9.0

WMIP: A Working-Memory-Inspired Programmable Probe of Cumulative State Tracking in Large Language Models

2026-10-30 · Dengzhe Hou, Lingyu Jiang, Deng Li, Zirui Li, Fangzhou Lin, Kazunori D Yamada

General AI

Human working memory lets people maintain and update information. Large language models (LLMs) face a related demand when tracking updates within a single prompt without writing out intermediate states, yet existing tests differ in their constraints on written steps. We introduce a working-memory-inspired probe of cumu…

Review
pending
Role
unreviewed
Read
soon
huggingface Score 7.4

Interactive Visual Gym: Evaluating Visual Interaction Models with Simulated Users in Interactive Worlds

2026-11-05 · Chenhao Zheng, Zixian Ma, Kai Li, Kanchana Ranasinghe, Jieyu Zhang, Artemis Panagopoulou, Weikai Huang, Jianxu Shangguan, Yuanyuan Tan, Silvio Savarese, Juan Carlos Niebles, Ranjay Krishna, Ran Xu, Honglu Zhou

General AI

Visual interaction models that support users during ongoing physical tasks require evaluation in closed loop, where assistant responses can change subsequent user behavior and task outcomes. Existing benchmarks largely rely on offline replay and therefore cannot capture these effects. We introduce Interactive Visual Gy…

Review
pending
Role
unreviewed
Read
soon
huggingface Score 6.4

Read Less, Solve More: Token-Efficient Sparse Reading of External Content for AI Agents

2026-11-07 · Zedong Liu, Jiaan Wu, Xinyang Ma, Le Xu, Kai Wang, Yuanchao Hu, Dingwen Tao, Guangming Tan

General AI

Web-facing AI agents repeatedly access webpages, documents, repositories, and tool responses, yet reading interfaces often expose entire objects when only sparse evidence is needed. This over-reading increases token and latency costs and can dilute relevant evidence. We present SparseRead, a reusable reading control la…

Review
pending
Role
unreviewed
Read
later
huggingface Score 6.4

Evolving Infra: Agent-Adaptive LLM Serving at Scale

2026-11-09 · Hao Kang, Weili Xu, Ziyang Li, Zheng Du, Hanchen Li, Tian Xia, Song Bian, Tushar Krishna, Huizi Mao, Ion Stoica, Song Han

General AI

Agentic workloads are a major source of LLM inference demand. Existing prefill--decode (PD) disaggregated serving systems offer limited flexibility for runtime adaptation of compute allocation and KV cache management, and rely on coarse signals that only partly capture the diverse and dynamic agent workload. These limi…

Review
pending
Role
unreviewed
Read
later
huggingface Score 6.4

Where Does Alignment Emerge and What Does It Mean? A Closer Look at the Platonic Representation Hypothesis

2026-11-09 · Julia Herbst, Simon Roschmann, Quentin Bouniot, Zeynep Akata

General AI

The Platonic Representation Hypothesis (PRH) suggests that neural networks are converging toward similar representation spaces. As language models become more capable, their representations are aligning with those of vision models. However, recent work has questioned whether this trend continues to hold for newer langu…

Review
pending
Role
unreviewed
Read
later
huggingface Score 6.0

Bits or permutations? Spending a fixed inference budget on weight precision or on option-order averaging in small local language models

2026-11-10 · Mohammed Nayeem Khan, Logesh L, Aliyas Ahmad Misri

General AI

Local language models can save memory by storing weights in fewer bits, and they can reduce sensitivity to the order of answer options by averaging a few reorderings. We ask whether the first makes the second worse, and which is the cheaper way to spend a fixed budget. On Qwen3-0.6B and Qwen3-1.7B with simulated round-…

Review
pending
Role
unreviewed
Read
later
huggingface Score 6.0

FAME-QP: A GPU-Accelerated Quadratic-Programming Solver for Real-Time Model Predictive Control

2026-11-10 · Santosh Mohan Rajkumar, Debdipta Goswami, Stephanie Stockar

General AI

We present FAME-QP, a GPU-accelerated solver for dense quadratic programs arising in linear model predictive control. FAME-QP combines constraint-dependent operator representations with fused single- and two-kernel iterations replayed through CUDA graphs, automatically selecting the execution configuration through devi…

Review
pending
Role
unreviewed
Read
later
huggingface Score 6.0

PolyRoute: A Twelve-Language and Code-Mixed Benchmark for Typed-Decision Routing, with Baselines and a Leaderboard

2026-11-10 · Mohammed Nayeem Khan, Sarun Kumar K S

General AI

Typed-decision encoders answer a question about a text in one forward pass, but their behaviour outside English is rarely measured on a fixed yardstick. We release PolyRoute: 322 native-language support messages in 12 languages and 9 writing systems, 156 code-mixed or romanised messages (Hinglish, Spanglish, Arabizi), …

Review
pending
Role
unreviewed
Read
later
huggingface Score 6.0

Precision on Demand: Escalating Weight Bits at Uncertain Decoding Steps in a Small Local Model

2026-11-10 · Mohammed Nayeem Khan, Logesh L, Aliyas Ahmad Misri

General AI

Local decoding usually fixes one weight precision for the whole generation. We study a different allocation: store 8-bit weights, decode with their top 4 bits, and redo a step at 8 bits only when the low-precision output distribution looks uncertain (entropy above a threshold or a small top-two margin). Quantization is…

Review
pending
Role
unreviewed
Read
later
huggingface Score 5.4

Module-Aware Optimization for Graph Neural Networks

2026-11-05 · Guy Hadad, Haggai Roitman, Moshe Eliasof, Bracha Shapira

General AI

Different optimizers impose different geometries on parameter updates. Adaptive methods such as Adam and AdamW use coordinate-wise updates, while matrix-aware optimizers such as Muon utilize weight-matrix structure through orthogonalized updates. For Graph Neural Networks (GNNs), however, the effective geometry of its …

Review
pending
Role
unreviewed
Read
later