arXiv subjects

Mingkai Zheng

Publications and source records attributed to Mingkai Zheng.

At least 19 recordsLinked to original sources

arXiv2026

A State Change Is Local, Its Consequences Are Not: Dependency-Aware State Revision for Long-Term Agent Memory

Long-lived agents rely on persistent memory of users, tasks and decisions, and this memory must stay consistent with a world that keeps changing. When an upstream fact changes, however, memory systems can revise the contradicted memory itself while the beliefs, plans and actions derived from it remain accepted, even though they have lost their support. We therefore formulate dependency-aware state revision, which asks not only what changed but also which dependent states should no longer be trusted. To address it, StateGraph maintains an evolving state dependency graph whose edges are recovered from interaction histories. Its core design is counterfactual verification: each candidate edge is tested by asking whether the dependent state would survive the loss of its premise, which separates support dependence from semantic relatedness, and revision then invalidates a dependent state only when a strict dependency fails and no independent support path survives. To test this design, we introduce StateChangeBench, a controlled benchmark of 50 author-constructed cases annotated with cascade depth, hard negatives and action consequences. On it, StateGraph's Impact-F1 margin over a CUPMem reimplementation is 6.9 percentage points at depth 1 and 26.0 at depth 3+. On public long-term memory benchmarks, its point estimates are comparable to or higher than those of the evaluated baselines on most datasets. These results provide initial evidence that explicit dependency tracking can improve selective memory revision under changing conditions.

cs.CL↗

arXiv2026

When Does Execution Provenance Help Agent Memory Retrieval?

A language agent's execution history can exceed its context window, requiring its memory system to retrieve complete supporting evidence under a hard token budget. Evidence may span multiple execution events, yet conventional retrievers use fixed token windows and fixed-k metrics that reward individual fragments without showing whether the complete evidence set fits in context. Smaller windows reduce irrelevant text but scatter evidence across candidates, while flat-versus-graph comparisons can conflate candidate design with graph propagation. To address these limitations, we formulate agent-memory retrieval as budgeted evidence completion and score exact gold spans in shared source coordinates. We first construct source-aligned provenance units from tool arguments and outputs. We then apply a zero-initialized residual R-GCN to refine frozen dense-retrieval scores over typed provenance edges. We evaluate 2,000 span-grounded memory queries over 1,207 held-out execution-grounded ISETrace trajectories. With matched Dense-FT scoring, provenance units improve Full Support@2048 by 19.07 points over flat 512-token windows and remain 11.96 points above a per-metric oracle over four flat chunk sizes; the pattern also holds with cross-encoder scoring. Holding the candidates and seed scores fixed, graph propagation adds 4.55 points in Full Support@2048 (95% CI [2.98, 6.18]). This gain is concentrated when gold evidence spans multiple events; entity co-occurrence expansion produces no comparable benefit, and relation and topology controls confirm dependence on typed transformations and observed graph structure. Overall, source-aligned candidates address the dominant granularity trade-off, while graph-conditioned propagation adds a smaller, targeted benefit for distributed evidence.

cs.MA↗

arXiv2026

Learning What to Ask for Tomorrow: Active Preference Acquisition for Long-Term Personalized Web Agents

Web agents increasingly serve the same user across repeated online tasks such as booking travel, shopping, and scheduling, so their long-term usefulness depends on how well they know the user's preferences. Passive memory and personalization methods cannot obtain preferences the user never reveals, and clarification methods that value a question by its benefit to the current task overlook the benefit of reusing the answer in later tasks. We study Active Preference Acquisition (APA): which missing preferences are worth the cost of asking? We define the Prospective Personalization Value (PPV) of a missing preference as its expected cumulative utility gain over future tasks minus the effort and disclosure-risk cost of asking, and select questions under a per-interaction budget. We characterize when cumulative reuse justifies acquisition despite insufficient single-task benefit, and evaluate this regime with an initial-profile diagnostic. APA computes the counterfactual gain of each candidate preference on public task prototypes with the agent's own task solver and aggregates it by the expected reuse of each task family. We evaluate it on ActiveWebPref, a controlled longitudinal benchmark with hidden but queryable user profiles and exact task utilities. At a horizon of 20 tasks, 34.4% of budget-feasible valuable missing preferences are amortization-dependent; APA acquires 80.3% of them, whereas current-task value of information acquires 5.1%. Under per-user asking-cost caps matched to APA's realized expenditure, APA achieves higher net utility than four heuristic selection baselines. In our cost setting, asking about every unknown preference yields lower net utility than not asking, while APA recovers 84.2% of its utility gain with 27.6% of its questions. The value of acquiring a user preference should account for its reuse across future tasks, not only its benefit to the task at hand.

cs.CL↗

arXiv2026

From Faulty Memories to Corrected Actions: Dependency-Guided Rollback Repair for Memory-Augmented Agents

Persistent memory lets language model agents reuse information across sessions, but it also makes errors durable: a poisoned, stale, or misattributed record can alter reasoning, tool use, answers, and subsequent memory writes. Existing defenses mainly detect or delete suspicious memories, or revise the current response. Deleting the source leaves already propagated claims, actions, and derived memories active, whereas resetting the store or replaying the full trace destroys benign state and repeats unnecessary computation. We therefore formulate post-failure memory recovery: given a failed execution and diagnosed faulty memories, regenerate the answer and clean up affected persistent state while retaining unaffected work. Our dependency-guided rollback repair builds a typed memory-to-action graph from runtime provenance, traces explicit downstream dependencies, preserves candidates with independent trusted support, deactivates unsupported memory state, and selectively replays only answer-relevant affected computation. We evaluate this approach on a 150-case controlled benchmark spanning three tool-use domains and four memory failure types, and on a 50-case trajectory-derived stress test adapted from LongMemEval-V2. On the controlled benchmark, it achieves 85.3% immediate recovery versus 77.3% for LLM-judge repair, the strongest baseline, while preserving all benign memories and using 41.8% fewer LLM calls. Selective replay, however, leaves higher recurrence than LLM-judge repair (26.6% vs. 12.1%), exposing a trade-off between targeted repair and broader recomputation. On the adapted subset, it reaches 68.0% recovery versus 54.0% for the next best method, with 5.9% recurrence. Code and data are available at https://github.com/xiaoqi-7/memory-rollback.

cs.AI↗

arXiv2026

Memories That Outlive Their Reasons: Dependency-Tethered Memory for Tool-Using LLM Agents

Agents now act on the Web, publishing packages, models and datasets through registry APIs, and remembering past verifications spares them the work. But a verdict rests on premises. When a signing key is revoked, everything built on it stops passing, yet the advisory names only the key, and the old verdicts of its dependents may never share a retrieval window with it. This non-locality of invalidation turns memory into a liability: in long simulated chats, an agent reading a memory of atomic facts publishes 62% of the objects whose failure lies upstream, each backed by a record that was true when written. Yet derivation is local even when invalidation is not. Tether records, at write time, the exact versions each verdict derives from, and at read time withdraws any verdict whose support has failed anywhere in the store; the agent reads the same fields as before. On Fallout, a generator of versioned worlds with hypothesis-blind simulated histories shared by every memory, Tether publishes 0.3% of upstream-failing objects across three domains when relations are stated with their verdicts, with 37% fewer verification calls than checking everything. A-Mem and Graphiti re-verify most candidates (0.83--0.86 checks per candidate), which limits their savings, yet publish 9% of upstream-failing objects; Mem0 relies on memory more and publishes 42%. Flushing every verdict on any failure is as safe as Tether but costs 32% more checks. The gain over fact memory holds with three open-weight backbones, a whole-store window and real Go advisories. These gains depend on capturing dependencies: when relations arrive in sessions apart from their verdicts, Tether's downstream unsafe publication rate rises to 9.4--22.1%. It also degrades with weaker writers that miss dependencies or invalidation events.

cs.CL↗

arXiv2026

When Is a Memory Safe Enough to Act On? Action-Conditioned Verification for Long-Lived Agents

Long-lived agents act on memory carried across sessions, and many of their actions are irreversible, so a stale memory can turn directly into harm. Existing work improves the memory itself --- retrieving, refreshing, or reconciling entries --- yet whether an entry must be re-checked before acting is not a property of the entry: the same remembered address can be used unchecked for a routine shipment but should be re-confirmed before a valuable one. We therefore study \emph{action-conditioned memory verification}: given a candidate action, decide which of its memory premises to re-verify and when to proceed or abstain, balancing execution risk, abstention, and verification cost. We formulate this as a sequential stopping problem, in which a verification is worth its cost only if it can change what the agent does next; under suitable verifier reliability and cost conditions, increasing action consequence can move the same memory through proceed, verify, and direct-abstention regimes. Building on this formulation, \textsc{AMV-Guard} is a practical verification gate that extracts the premises a candidate action rests on, calibrates beliefs about whether they are wrong, and selects checks by bounded value of information (VOI). On ToolSandbox, relative to a static policy given the same inputs, it lowers mean verification cost by 33.9\% while keeping 100\% safe-proceed recall and 0\% Gate harmful, and raises Proceed coverage from 38.83\% to 44.17\%; in a gate-core transfer to AppWorld, which carries over the stopping/VOI core but not the full structured pipeline, it lowers verification cost by 21.3\% at a small cost in safe-proceed recall and task success. These results suggest that verification effort should be set by what an agent is about to do, not only by how reliable a memory appears.

cs.MA↗

arXiv2026

From Agent Traces to Trust: A Survey of Evidence Tracing and Execution Provenance in LLM Agents

Large language model (LLM)-based agents are moving beyond text generation to planning, tool use, retrieval, memory access, environmental interaction, and multi-agent collaboration. Because their outputs depend on information and execution steps across an agent run, their execution processes are difficult to verify, debug, and audit. Even a correct final answer does not reveal which evidence supported its claims, whether tool calls were justified, how memory shaped decisions, or where execution failures originated. Relevant research addresses these questions across evidence attribution, tool-use safety, agent memory, and observability, but these areas emphasize different parts of the execution process. This survey brings them into a common framework through execution provenance, a typed graph of an agent run, and evidence tracing, its projection onto evidence-support relations, distinguishing evidence that supports a conclusion from execution that influences it. We introduce a six-dimensional taxonomy covering trace sources, evidence and execution units, provenance relations, tracing granularity and timing, representation forms, and trust functions, complemented by operational definitions of nine provenance relations and two cross-cutting attributes, provenance origin and trace assurance. Applying this framework to the surveyed systems and benchmarks reveals three gaps: limited integration of fine-grained provenance with multiple trust functions, no benchmark in our catalog that evaluates the complete provenance chain, and little reporting of tamper-evident or attested traces. These findings motivate research on unified trace schemas, semantic provenance, memory and multi-agent lineage, provenance-aware safety and recovery, realistic benchmarks, trace assurance, and privacy-aware auditing. Resources are available at https://github.com/xiaoqi-7/Agent-Tracing-Survey.

cs.CR↗

arXiv2026

IndustrialVLA-Bench: A Traceable Multi-Axis Evaluation of Open Robot Policy Models

Open robot policies increasingly follow two paradigms: vision-language-action models (VLAs) directly map observations and instructions to actions, whereas world-action models (WAMs) incorporate learned video or world dynamics into policy learning or action generation. Although both target the same manipulation tasks and represent alternative design choices, they are commonly reported under different evaluation protocols, leaving their capability, robustness, language sensitivity, and deployment-cost trade-offs unclear. We present IndustrialVLA-Bench, an evidence-aware evaluation of six released VLA and WAM systems under a unified reporting schema. It separately evaluates clean capability on LIBERO, non-language robustness on LIBERO-Plus, instruction sensitivity on LIBERO-Para, and observed execution cost. Reported task scores aggregate three complete evaluations with distinct random seeds under a fixed checkpoint and inference configuration. Across all six systems, clean LIBERO averages differ by only 1.58 points, whereas robustness and paraphrase summaries span 14.62 and 31.08 points. Restricting every comparison to the three protocol-faithful systems preserves the effect (1.36, 14.62 and 23.10 points), so the diagnostic separation reported here does not depend on the weaker evidence tiers. We additionally report observed inference latency, peak memory, runtime mode, and an evidence status for every system. Protocol-faithful, near-reproduction, and pending-verification entries remain visibly separated; only protocol-faithful entries support strict comparisons. Rather than claiming universal superiority of either paradigm, IndustrialVLA-Bench provides traceable evidence for comparing released robot policies on shared practical criteria. Code and evaluation records are available at https://github.com/xiaoqi-7/IndustrialVLA-Bench.

cs.RO↗

arXiv2026

PACMI: Provenance-Aware Cascading Memory Invalidation for Long-Term LLM Agents

LLM agents rely on long-term memory to retain and reuse information when performing tasks over long horizons. Existing methods provide limited support for handling memories that become outdated as new observations or domain evidence arrive. Such outdated memories may remain semantically relevant, continue to affect dependent records, and retain value as historical evidence. This calls for two capabilities: dependency tracking to identify downstream effects and historical preservation to retain useful past records. We propose Provenance-Aware Cascading Memory Invalidation (PACMI), a framework that represents memories and new evidence in a provenance graph with typed dependency edges. PACMI assigns records to a four-state validity lattice, propagates validity changes to dependent memories, and uses the resulting states for retrieval and stale-premise detection. We also introduce a diagnostic benchmark with 100 cases and 300 queries across five domains. The evaluation separates node, context-, and answer-level performance. PACMI achieves the highest final-answer accuracy on this benchmark, and its paired difference from the strongest baseline is significant under an exact McNemar test. The premise checker achieves perfect precision, recall, and F 1 on the controlled query distribution. Cascading propagation primarily improves memorystate correctness: removing it increases final-answer errors from 3 to 11, but the paired difference does not reach the 0.05 significance threshold. Code and data will be made publicly available.

cs.LG↗

arXiv2026

Broken Symmetry in BF16 Attention: Why FlashAttention Gradients Blow Up Late in Training

BF16 is now standard in large-scale pretraining, including in fused attention kernels such as FlashAttention, and these kernels are widely trusted. When we used FlashAttention-3 to pretrain a 450M-parameter transformer on 50B tokens, however, we ran into a problem: training was healthy for 25B tokens, then the gradient norm grew a thousandfold and the loss ended 0.2 nats above FP32 attention, without a single NaN. Recomputing the attention backward of just two layers in FP32 removes almost all of the excess gradient. Part of the cause is known: a fused multiply-add in the forward softmax, so far treated as an extreme-input NaN case and never fixed in FlashAttention-3. Repairing it stops the blow-up, but the query gradient is still wrong by more than its own size, and training still drives attention logits to thousands of times their size under accurate gradients. The remaining error comes from a broken conservation law. The softmax score gradient sums to zero along every row, which makes the query gradient blind to where the keys sit as a group; rounding it to BF16 leaves a small nonzero sum that leaks the mean key into the gradient, and the leak grows exactly as late training makes keys large and attention sharp. We introduce GProj (gauge projection), which restores the zero sum after the cast with two rank-one corrections per row. It cuts the remaining median query/key gradient errors from 219%/13% to 0.34%/0.37%, on par with FP32 attention, for 4.7% more time per training step. In matched from-scratch runs it trains to the same loss as FP32 attention, while FlashAttention-3 and key smoothing both destabilize.

cs.LG↗

arXiv2026

Diamond Agent: Agentic Control of Federated HPC Resources as a Service

Efficiently aggregating and orchestrating computing power across heterogeneous clusters for HPC workflows faces four practical challenges: preserving workflow context across independently administered clusters, moving large datasets between sites, reasoning about site-specific environments and scheduler policies, and exploiting live queue and resource states for efficient task scheduling. To this end, we design Diamond Agent, an agentic system that enables intelligent execution of HPC workflows across heterogeneous clusters with typed skills as the interface. Diamond Agent provides an agent-facing workspace and skills that unify cross-site resource discovery, resource specification, data movement, task execution, and result retrieval. A centralized Diamond Agent instance can operate multiple supercomputers without being deployed separately on each login node. Diamond Agent translates high-level agent actions into valid site-specific executions, moves data through Globus Transfer, and uses live system capability and queue information to select feasible placements. Its event-driven continuation mechanism decouples agent actions from long-running batch jobs: persistent services monitor remote execution and resume the agent only when a result or decision-relevant event is available. We experiment with 27 hours of telemetry and 19 matched multi-site submission rounds comprising 83 jobs across four production supercomputers. Compared with a fixed-site baseline, Diamond Agent reduces the median additional completion time relative to the fastest observed placement from 42 seconds to 4 seconds, a 10.5x reduction.

cs.DC↗

arXiv2026

ReBridge-Flow: Re-Coupling Posterior Bridges in Flow Matching for Image Restoration

Flow Matching provides an efficient generative prior for image restoration by learning continuous transport between source and data distributions. However, existing methods typically incorporate measurement constraints through local corrections. Such corrections may disrupt the source-clean endpoint coupling implicitly encoded by the pretrained flow, making the corrected endpoint pair incompatible with the current state. To address this issue, we propose ReBridge-Flow, a posterior bridge re-coupling method. Specifically, given the current state, ReBridge-Flow first decodes the corresponding local source and clean endpoints. It then incorporates measurement information through clean-side anchoring and synchronously re-couples the source endpoint, yielding a measurement-aware endpoint pair with improved local bridge compatibility. The re-coupled endpoints further define a posterior-informed transport direction for advancing the sampling process. We also introduce the Posterior Bridge Defect, which jointly characterizes measurement error, deviation from the flow prior, and bridge mismatch, and leads to explicit updates for clean-side anchoring and source-side re-coupling. Extensive experiments on multiple natural and medical image restoration tasks demonstrate that ReBridge-Flow effectively alleviates bridge mismatch and improves the structural consistency of restored images.

cs.CV↗

arXiv2026

MAP-Graph: Provenance-Aware Shared Memory for Multi-Agent Workflows

Shared memory helps language-model agents reuse information across long workflows, yet relevant evidence may not be admissible for a particular agent or action. Because restrictions propagate through derivations, summaries can conceal private, poisoned, untrusted, or revoked sources, enabling unauthorized reads or unsafe actions. Existing approaches provide semantic retrieval, scoped access, or lineage tracking, but do not clearly separate hard authorization from graded trust or adapt evidence requirements to action risk. We introduce MAP-Graph, a provenance-aware memory layer that represents agents, sources, memories, claims, and actions in a typed execution graph. It traces ancestry, excludes permission-ineligible records, reranks eligible memories by semantic similarity and multiplicative path trust, and applies a risk-sensitive gate before action execution while retaining affected lineage for audit. On a controlled benchmark of 2,700 synthetic tasks per method across three domains, MAP-Graph achieves 94.96\% overall task success, 72.70\% exact decision accuracy, and 90.22\% in the clean setting, where success requires a correct \textsc{Allow} rather than a safe intervention. Ablations isolate the roles of permission filtering, path trust, and action gating, while transfer tests with two additional backbones preserve the exact-decision and access-control advantages. These results support provenance as an operational control signal, rather than only post-hoc audit metadata, within the evaluated setting.

cs.AI↗

arXiv2026

MemTX: Transactional Belief Commit for Stateful Agent Memory

LLM agents increasingly coordinate through persistent shared memory: one agent's write becomes another agent's premise, and eventually a tool call with real side effects. Current agent memory systems treat every accepted write as immediately actionable truth, so a polluted tool result, a stale update, or a teammate's half-finished note can silently drive an irreversible action. We argue that a memory write is not a belief commit. We present MemTX, a transactional belief-commit protocol. Each record carries evidence, permissions, provenance, and validity. Writes are staged inside snapshot-isolated transactions and admitted by a validate-and-commit pipeline, irreversible tool calls are gated on in-flight belief state, and retracting a belief triggers typed cascading repair of its derived records and tool side effects. Two invariants, action-safety gating and cascade-repair completeness, are machine-checked by property-based testing and bounded exhaustive enumeration of 5.5 million protocol states, with zero violations. Across five backbones from three model families, MemTX leads all eight baselines with paired-McNemar significance on four backbones and statistically ties the best baseline on the fifth and strongest, while remaining the only method with zero downstream harm on every backbone. Backbone capability does not substitute for commit discipline.

cs.AI↗

arXiv2026

PHOENIX: Resilient LLM Training with Hot-Swapping via Zero-Overhead Checkpoint

State-of-the-art large language model (LLM) training takes tens of thousands of graphics processing units (GPUs) for months and encounters failures across the software and hardware stack. Existing fault-tolerance mechanisms either impose non-trivial overhead during failure-free execution or suffer from prolonged recovery latency, particularly under scenarios where a small subset of compute nodes experience permanent failures. %The tradeoff between failure-free overhead and recovery latency forms a space forms a Pareto frontier We present PHOENIX to simultaneously address both optimization objectives. PHOENIX incorporates a fault-tolerance mechanism that restores LLM training via hot-swapping, namely by replacing failed nodes with spare nodes without terminating the complete job. The hot-swapping of PHOENIX is enabled by two ideas: First, it exploits an off-critical-path in-memory checkpointing mechanism for spatial redundancy. Second, it introduces a communicator reconstruction protocol that replaces failed nodes with spare nodes at runtime. PHOENIX efficiently overlaps the in-memory checkpointing with computation, thus introducing zero overhead during error-free execution. Upon permanent node failures, PHOENIX can rebuild memory states with minimal recomputation by leveraging in-memory checkpoints. We evaluate PHOENIX across scales (up to 512 NVIDIA A100 GPUs) and LLMs (up to 65B parameters), and observe zero checkpoint overhead with hot-swapping recovery completing in under 40 seconds. These results show that PHOENIX simultaneously achieves both zero-overhead error-free execution and extremely low recovery cost.

cs.LG↗

arXiv2026

SCAPE: Accurate and Efficient LLM Training with Extreme Sparse Communication

Communication increasingly dominates the cost of Large Language Model (LLM) pre-training, especially under data-parallel and sharded training schemes, where gradient synchronization and parameter reconstruction overhead increase with model size and system scale. Existing communication-reduction methods either sparsify raw gradients, which can be unstable for modern Adam-style optimizers at high sparsity, or quantize communication, whose savings are fundamentally bounded by bit width and often incur additional runtime overhead. We present SCAPE, a communication-efficient distributed optimizer for LLM training that exploits the stability of AdamS's first-moment to enable aggressive sparsification without loss of LLM quality. Instead of constructing masks from raw gradients, SCAPE derives them from first-moment-based statistics, partitions mask generation across workers to align with optimizer sharding, and delays mask usage by one step so that mask synchronization can overlap with computation. SCAPE also reconstructs the quantities required for second-moment updates from a single synchronized sparse buffer, avoiding an additional collective. We implement SCAPE in Megatron-LM and evaluate its convergence by pre-training GPT-345M on OpenWebText and Llama-500M on SlimPajama-6B using 32 NVIDIA GH200 GPUs on TACC Vista. In both models, SCAPE preserves training stability, validation loss, and downstream task accuracy under 90\% and 99\% sparsity. For Llama-500M, SCAPE reduces end-to-end pre-training wall-clock time by up to 43.3\% while maintaining model quality comparable to dense AdamW and AdamS. For Llama-1.8B, SCAPE achieves up to 3.26$\times$ speedup per step compared to dense AdamS.

cs.LG↗

arXiv2025

Efficient Fine-Grained GPU Performance Modeling for Distributed Deep Learning of LLM

Training Large Language Models(LLMs) is one of the most compute-intensive tasks in high-performance computing. Predicting end-to-end training time for multi-billion parameter models distributed across hundreds of GPUs remains challenging due to complex interactions between transformer components, parallelism strategies(data, model, pipeline, tensor), and multi-tier communication. Learned models require costly sampling, while analytical models often struggle with real-world network and hardware complexities. We address this by decomposing LLMs into core computational primitives and modeling them with: (1) operator-level decomposition for fine-grained analysis; (2) lightweight sampling based hardware-aware prediction models for key operations; (3) an end-to-end prediction system integrating these components across complex parallelization strategies. Crucially, our methodology has been validated on two large-scale HPC systems. Our framework achieves low average prediction errors-4.98\% on Perlmutter(A100) and 9.38\% on Vista(GH200)-for models up to 20B parameters across 128 GPUs. Importantly, it runs entirely on CPUs, enabling rapid iteration over hardware configurations and training strategies without costly on-cluster experimentation.

cs.DC↗

arXiv2025

Towards Analyzing and Understanding the Limitations of VAPO: A Theoretical Perspective

The VAPO framework has demonstrated significant empirical success in enhancing the efficiency and reliability of reinforcement learning for long chain-of-thought (CoT) reasoning tasks with large language models (LLMs). By systematically addressing challenges such as value model bias, heterogeneous sequence lengths, and sparse reward signals, VAPO achieves state-of-the-art performance. While its practical benefits are evident, a deeper theoretical understanding of its underlying mechanisms and potential limitations is crucial for guiding future advancements. This paper aims to initiate such a discussion by exploring VAPO from a theoretical perspective, highlighting areas where its assumptions might be challenged and where further investigation could yield more robust and generalizable reasoning agents. We delve into the intricacies of value function approximation in complex reasoning spaces, the optimality of adaptive advantage estimation, the impact of token-level optimization, and the enduring challenges of exploration and generalization.

cs.LG↗